Joint Determination of Reactor Antineutrino Spectra from ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$ Fission by Daya Bay and PROSPECT

A joint determination of the reactor antineutrino spectra resulting from the fission of ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$ has been carried out by the Daya Bay and PROSPECT Collaborations. This Letter reports the level of consistency of ${}^{235}\mathrm{U}$ spectrum measurements from the two experiments and presents new results from a joint analysis of both data sets. The measurements are found to be consistent. The combined analysis reduces the degeneracy between the dominant ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$ isotopes and improves the uncertainty of the ${}^{235}\mathrm{U}$ spectral shape to about 3%. The ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$ antineutrino energy spectra are unfolded from the jointly deconvolved reactor spectra using the Wiener-SVD unfolding method, providing a data-based reference for other reactor antineutrino experiments and other applications. This is the first measurement of the ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$ spectra based on the combination of experiments at low- and highly enriched uranium reactors.

Figure 1

Top: Reconstructed prompt energy distributions based on simulated ${\overline{\nu }}_e$ signals with specific ${\overline{\nu }}_e$ energy ranges (uniform distribution). The areas of the distributions are normalized to 1. The shift in peak location between the two experiments is driven primarily by the handling of scintillator nonlinearity in the energy response. These effects for PROSPECT are incorporated into the response function while this effect is taken into account by Daya Bays calibration methodology. Bottom: FWHM of reconstructed prompt energy distributions versus prompt energy at the peak of those distributions. The difference in FWHM is primarily due to various effects from inactive volume in the detector response functions.

Figure 2

Comparison of the measurements of the ${}^{235}\mathrm{U}$ prompt energy spectrum from Daya Bay and PROSPECT (top) and the ratio of the spectrum from Daya Bay over the one from PROSPECT (bottom). Here, the measurement from Daya Bay has been transformed to the reconstructed energy scale of PROSPECT based on a dedicated response matrix ${\mathit{R}}^{\mathrm{map}}$ and the $y$ axis has been scaled to match the absolute rate from the Daya Bay measurement. Error bars contain both statistical and systematic contributions. The measurements from Daya Bay and PROSPECT are consistent with each other.

Figure 3

Top: the extracted ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$ spectra in Daya Bay’s prompt energy from the combined analysis of the Daya Bay and PROSPECT data. The corresponding scaled Huber model predictions are overlaid. The error bars in the data points are the square root of the diagonal elements of the covariance matrix for the spectral shape, with no absolute rate uncertainty. Bottom: the ratio of the combined analysis results to the shape predictions from the scaled Huber-Mueller model.

Figure 4

Top: the ratio of the combined analysis results to the Daya Bay only results [50]. Middle: the difference of the relative uncertainties between the combined analysis results and the Daya Bay only results [50]. The inset shows the enlarged plot of the relative uncertainty differences. Bottom: the correlation coefficients of the extracted prompt energy spectra between ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$.

Figure 5

Top: ${}^{235}\mathrm{U}$ and ${}^{239}\mathrm{Pu}$ antineutrino spectra unfolded from the jointly deconvolved Daya Bay and PROSPECT measurements. Bottom: ratio of the measurements to their respective models, which are corrected by the smearing matrices ${\mathit{A}}_c$ in both panels.