We report on a new open-source, user-friendly numerical relativity code package called SENR/NRPy+. Our code extends previous implementations of the BSSN reference-metric formulation to a much broader class of curvilinear coordinate systems, making it ideally suited to modeling physical configurations with approximate or exact symmetries. In the context of modeling black hole dynamics, it is orders of magnitude more efficient than other widely used open-source numerical relativity codes. NRPy+ provides a Python-based interface in which equations are written in natural tensorial form and output at arbitrary finite difference order as highly efficient C code, putting complex tensorial equations at the scientist's fingertips without the need for an expensive software license. SENR provides the algorithmic framework that combines the C codes generated by NRPy+ into a functioning numerical relativity code. We validate against two other established, state-of-the-art codes, and achieve excellent agreement. For the first time-in the context of moving puncture black hole evolutions-we demonstrate nearly exponential convergence of constraint violation and gravitational waveform errors to zero as the order of spatial finite difference derivatives is increased, while fixing the numerical grids at moderate resolution in a singular coordinate system. Such behavior outside the horizons is remarkable, as numerical errors do not converge to zero near punctures, and all points along the polar axis are coordinate singularities. The formulation addresses such coordinate singularities via cell-centered grids and a simple change of basis that analytically regularizes tensor components with respect to the coordinates. Future plans include extending this formulation to allow dynamical coordinate grids and bispherical-like distribution of points to efficiently capture orbiting compact binary dynamics.

We solve the Hamiltonian and momentum constraints of general relativity for two black holes with nearly extremal spins and relativistic boosts in the puncture formalism. We use a non-conformallyflat ansatz with an attenuated superposition of two Lorentz-boosted, conformally Kerr or conformally Schwarzschild 3-metrics and their corresponding extrinsic curvatures. We compare evolutions of these data with the standard Bowen-York conformally flat ansatz (technically limited to intrinsic spins χ = S/M 2 ADM = 0.928 and boosts P/MADM = 0.897), finding, typically, an order of magnitude smaller burst of spurious radiation and agreement with inspiral and merger. As a first case study, we evolve two equal-mass black holes from rest with an initial separation of d = 12M and spins χi = Si/m 2 i = 0.99, compute the waveforms produced by the collision, the energy and angular momentum radiated, and the recoil of the final remnant black hole. We find that the black-hole trajectories curve at close separations, leading to the radiation of angular momentum. We also study orbiting nonspinning and moderate-spin black-hole binaries and compare these with standard Bowen-York data. We find a substantial reduction in the nonphysical initial burst of radiation which leads to cleaner waveforms. Finally, we study the case of orbiting binary black-hole systems with spin magnitude χi = 0.95 in an aligned configuration and compare waveform and final remnant results with those of the SXS Collaboration [1], finding excellent agreement. This represents the first moving punctures evolution of orbiting and spinning black holes exceeding the Bowen-York limit. Finally, we study different choices of the initial lapse and lapse evolution equation in the moving punctures approach to improve the accuracy and efficiency of the simulations.

We use fully nonlinear numerical relativity techniques to study high energy head-on collision of nonspinning, equal-mass black holes to estimate the maximum gravitational radiation emitted by these systems. Our simulations include improvements in the construction of initial data, subsequent full numerical evolutions, and the computation of waveforms at infinity. The new initial data significantly reduces the spurious radiation content, allowing for initial speeds much closer to the speed of light, i.e. v ∼ 0.99c. Using these new techniques, We estimate the maximum radiated energy from head-on collisions to be Emax/MADM = 0.13 ± 0.01. This value differs from the second-order perturbative (0.164) and zero-frequency-limit (0.17) analytic computations, but is close to those obtained by thermodynamic arguments (0.134) and by previous numerical estimates (0.14 ± 0.03).