Uncertainty-quantified phenomenological optical potentials for single-nucleon scattering

Optical-model potentials (OMPs) continue to play a key role in nuclear reaction calculations. However, the uncertainty of phenomenological OMPs widespread use---inherent any parametric model trained on data---has not been fully characterized, and its impact downstream users remains unclear. Here we assign well-calibrated uncertainties for two representative global OMPs, those Koning-Delaroche Chapel Hill '89, using Markov-chain Monte Carlo parameter inference. By comparing canonical versions these against experimental data originally used constrain them, show how lack outlier rejection systematic underestimation contributes bias of, overconfidence in, best-fit values. Our updated, uncertainty-quantified address issues yield complete covariance information potential parameters. Scattering predictions generated from our ensembles improved performance both original training corpora new ``test'' corpus comprising many single-nucleon scattering collected over last twenty years. Finally, apply case studies application-relevant cross sections. We conclude that, common applications including OMP should become standard practice. To facilitate their immediate use, digital updated related tools forward propagation are included as Supplemental Material.