Gravitational wave forms from the final plunge and coalescence of black-hole and neutron-star binaries are primary targets for LIGO and other gravitational wave interferometers. These wave forms can only be predicted by large-scale numerical computations. Important theoretical and algorithmic issues remain to be solved before successful simulations of compact binary mergers can be achieved. A code for evolving Einstein's equations in three spatial dimensions and with no special symmetries assumed is needed to produce binary coalescence simulations. This code must be stable and accurate. Achieving stability will require theoretical advances in understanding how evolution schemes interact with boundary conditions when the spacetime is evolving dynamically. Evolving neutron stars requires incorporating matter sources into the Einstein equation solver and developing stable and accurate methods for evolving the matter.
Thomas Baumgarte and Stuart Shapiro will construct such a code to solve the problem of coalescing compact binaries, combining theory and computation in a symbiotic fashion to make the most rapid progress possible toward the goal of producing gravitational wave forms.