Research in Scientific Computation and Mathematical Modeling

Combined with the development of novel algorithms, scientific computation has not only joined experiment and theory as one of the fundamental tools of investigation, but it has also altered the kind of experiments performed and expanded the scope of theory. The Center for Scientific Computation and Mathematical Modeling (CSCAMM) was created recently by the University of Maryland College Park (UMCP), as a ‘major impact’ initiative which aims to address the challenges offered by these scientific developments. The primary goal of the Center is to foster research activity that highlights the interplay between novel computational algorithms and mathematical modeling, as complementary avenues to theory and experiments. To stimulate such research activity, CSCAMM fosters collaborative research programs which involve UMCP scientists together with graduate students, visiting post-docs and world renowned scholars. CSCAMM sponsors an active visitors’ program through a series of workshops and conferences. Finally, CSCAMM supports an educational program with a series of tutorials, courses and active involvement with the new UMCP Applied Mathematics and Scientific Computation graduate program.
Contact: Professor Eitan Tadmor director@cscamm.umd.edu 301-406-0648

High-resolution Algorithms
Research interests include analysis of time-dependent problems governed by linear and nonlinear PDEs, the development of novel high-resolution algorithms for the approximate solution of these problems, and the interplay between the theory and computational aspects of such approximate methods. Topics of focused interest include construction and implementation of non-oscillatory schemes, the study of kinetic formulation of nonlinear PDEs, development of convergence theory for approximate solutions of nonlinear convection equations, numerical algorithms for detection of edges and accurate recovery of piecewise smooth data from its spectral information, and the study of Approximate solutions of incompressible Euler equations.
Contact: Professor Eitan Tadmor tadmor@cscamm.umd.edu: 301-405-0648

Computational MHD
Research includes the development of kinetic (phase space) algorithms for high performance computing, with an emphasis on Eulerian schemes and closure theory, with focus on direct numerical simulation of collisionless, magnetized plasma turbulence for first-principles simulation of turbulent transport in magnetic confinement fusion devices and turbulent heating and particle acceleration in astrophysical plasmas and the development of lightweight, portable, high-performance components for practical high performance parallel computing.
Contact: Professor Bill Dorland bdorland@umd.edu 301-405-1647.

Nonlinear Approximation Theory
Research interests focus on nonlinear approximation including adaptivity (-- based on variable knots for example), rational functions, and compressed wavelets decompositions.
Contact: Professor Ron DeVore devore@cscamm.umd.edu 301-405-1217.

Back to Top