Extending NRPyElliptic to Extreme-GR, Beyond-GR, and GRMHD Applications

Numerical relativity (NR) simulations are essential for self-consistently modeling compact binary mergers, whose observations form the heart of gravitational-wave science. To make these models computationally tractable, the equations of gravity and relativistic hydrodynamics are recast as initial value problems, thus relying upon the careful construction of initial data. NRPyElliptic is a flexible elliptic solver that sets up initial data for NR simulations, leveraging NRPy code generation to construct its hyperbolic relaxation method. I will report on ongoing efforts to extend NRPyElliptic's compatibility with new computational infrastructures and to implement more sophisticated mathematical formalisms. The solver has been refined to generate initial data for highly spinning black-hole binaries and in effective theories of gravity. The code now supports both single-patch curvilinear grids within BlackHoles@Home, as well as Cartesian AMR grids within the Einstein Toolkit framework. In addition, I will demonstrate the solver's flexibility with a GRMHD use case, in which we recast the inverse-curl operation as an elliptic equation to compute the vector potential from a specified magnetic field.