Combined Analysis of all Three Phases of Solar Neutrino Data from the Sudbury Neutrino Observatory

TL;DR

This work delivers a final, high-precision global analysis of all Sudbury Neutrino Observatory solar-neutrino data, incorporating a new particle-identification technique in Phase III to greatly reduce backgrounds. By parameterizing the $^8$B neutrino signal with an energy-dependent survival probability $P_{ee}(E_ u)$ and its day-night asymmetry, and performing a combined fit to all phases plus KamLAND data, the study delivers a refined flux measurement $\Phi_B$, tighter oscillation-parameter constraints, and a stringent bound on $\sin^2\theta_{13}$. The results strongly support the LMA solution, yielding $\Delta m^2_{21}$ in the $\sim ext{few} \times 10^{-5}\,\text{eV}^2$ range and a nonzero but small $\theta_{13}$, with a 95% CL upper limit of $\sin^2\theta_{13} < 0.053$. Overall, the analysis demonstrates the power of combining multi-phase data with improved background rejection to achieve high-precision solar-neutrino oscillation measurements and robust cross-experiment consistency.

Abstract

We report results from a combined analysis of solar neutrino data from all phases of the Sudbury Neutrino Observatory. By exploiting particle identification information obtained from the proportional counters installed during the third phase, this analysis improved background rejection in that phase of the experiment. The combined analysis resulted in a total flux of active neutrino flavors from 8B decays in the Sun of (5.25 \pm 0.16(stat.)+0.11-0.13(syst.))\times10^6 cm^{-2}s^{-1}. A two-flavor neutrino oscillation analysis yielded \Deltam^2_{21} = (5.6^{+1.9}_{-1.4})\times10^{-5} eV^2 and tan^2θ_{12}= 0.427^{+0.033}_{-0.029}. A three-flavor neutrino oscillation analysis combining this result with results of all other solar neutrino experiments and the KamLAND experiment yielded \Deltam^2_{21} = (7.41^{+0.21}_{-0.19})\times10^{-5} eV^2, tan^2θ_{12} = 0.446^{+0.030}_{-0.029}, and sin^2θ_{13} = (2.5^{+1.8}_{-1.5})\times10^{-2}. This implied an upper bound of sin^2θ_{13} < 0.053 at the 95% confidence level (C.L.).

Figures (19)

• Figure 1: Schematic diagram of the SNO detector. We used a coordinate system with the center of the detector as the origin, and $z$ direction as vertically upward.
• Figure 2: $T_{\rm eff}{}$ spectra for the PMT background events obtained from a bifurcated analysis of data from Phase I including the best fits to Equation \ref{['eqn:PMTPDF']}.
• Figure 3: Sample waveforms. The top plot shows a neutron waveform (black) obtained from ${}^{24}{\rm Na}$ calibration data with the best fit to the neutron hypothesis (red). The bottom plot shows an alpha waveform (black) obtained from a string filled with ${}^{4}{\rm He}$ with the best fit to the alpha hypothesis (red). The vertical lines represent the fit boundaries.
• Figure 4: Distribution of particle identification parameters for neutron events (boxes, where the area represents the number of events) and alpha events (red marks). The line represents the boundary for cuts. PID cut 1 applies to parameters $p_a$ and $p_b$, and PID cut 2 applies to parameters $p_c$ and $p_d$ for events that failed PID cut 1.
• Figure 5: $E_{\rm NCD}{}$ spectrum before (brown) and after (red) the particle identification cut. From left to right the plots are for ${}^{24}{\rm Na}$ calibration data (neutrons), data from strings filled with ${}^{4}{\rm He}$ (alphas), and data from strings filled with ${}^{3}{\rm He}$.