Speaker
Description
The exact abundances of light elements produced during Big Bang Nucleosynthesis (BBN) are highly sensitive to the physics of the primordial plasma. Percent-level measurements of primordial deuterium have enabled precision cosmology from BBN; to match this level of precision in theoretical calculations, a careful estimate of uncertainties arising from nuclear reaction cross sections is of paramount importance. We present a novel method to incorporate measured nuclear cross section data directly into BBN calculations using Gaussian processes. With Gaussian processes, we generate a probability distribution over a broad class of functions consistent with the data without needing to choose a specific functional form for the fit, while fully accounting for both statistical and systematic uncertainties. We pass draws from the Gaussian process into the BBN code, propagating experimental uncertainties to robust predictions of primordial element abundances. We demonstrate the reliability of our method by generating realistic mock cross section datasets and consistently recovering the expected results within predicted uncertainties. Assuming the Planck value for the baryonic density of the universe ωb for our calculation of the primordial deuterium abundance, we find potentially important implications for the concordance between CMB and BBN within the ΛCDM model.