Surrogate modeling of binary black hole waveforms with rapid spins

Gravitational-wave signals from merging binary black holes carry information about their sources' masses and spins. Accurately recovering these parameters requires comparing the detected signals with many modeled waveforms. Numerical relativity (NR) produces the most accurate waveforms, but at a great computational cost. The technique of surrogate modeling allows for rapid construction of interpolated waveforms from a model trained with NR simulations. Prior SXS Collaboration surrogate models have not included black hole waveforms with spins greater than 0.8, although rapidly spinning black hole systems may be present in potential future observations. I will present an analysis of the performance of surrogate models for high spin black hole mergers. Specifically, I will evaluate the performance of one-dimensional models for equal mass and spin binary black hole systems in this parameter space, using mismatches with NR waveforms. I will also compare these results with the extrapolation of current NR surrogate models to high spin.