Informal (though at times slightly technical) descriptions of a few of my major research interests.
Before a numerical relativity simulation can begin, we need to turn the physical field equations into a well-posed differential equations problem for the computer.
Characterizing a black-hole collision requires us to define what we mean by the angular momentum of warped space.
It was long presumed that when two black holes collide, the merger remnant must settle down to a stationary black hole. Checking this required numerical relativity and some clever engineering.
Computational relativity codes provide their results as a sea of numbers. Translating that sea of numbers into a visceral picture of spacetime dynamics requires new techniques of scientific visualization.