Global Fit of KamLAND Data and the Daya Bay Antineutrino Energy Spectrum

This paper presents a global analysis combining KamLAND data with Daya Bay's independently measured fission antineutrino spectra, revealing that using these empirical spectra instead of the Huber-Müller model reduces the tension between KamLAND's and JUNO's measurements of solar neutrino oscillation parameters by lowering the best-fit values of $\Delta m^2_{21}$ and $\tan^2\theta_{12}$.

The Gist

The Big Picture: A Cosmic Puzzle with a Missing Piece

Imagine scientists are trying to solve a giant jigsaw puzzle about how tiny particles called neutrinos behave. These particles are like ghostly messengers that pass through everything, including the Earth.

For a long time, two different teams of scientists have been looking at the same puzzle pieces but seeing slightly different pictures:

1. The "Sun" Team (SNO/JUNO): They look at neutrinos coming from the Sun.
2. The "Reactor" Team (KamLAND): They look at neutrinos coming from nuclear power plants.

Both teams are trying to measure two specific numbers that describe how these particles "dance" (oscillate) as they travel:

• The Speed of the Dance ($\Delta m^2_{21}$): How fast the particles change their identity.
• The Angle of the Dance ($\theta_{12}$): How wide their steps are.

Recently, a new, very precise experiment called JUNO measured these numbers and found they were slightly different from what the KamLAND experiment found in 2013. It's like two people measuring the same table, but one says it's 100cm and the other says it's 100.2cm. They are close, but not quite matching.

The Suspect: A "Bumpy" Map

The author of this paper, Guihong Huang, suspects the problem isn't the neutrinos themselves, but the map the scientists are using to read them.

When the KamLAND team analyzed their data, they used a theoretical "map" (called the Huber-Müller model) to predict what the neutrino energy spectrum should look like. Think of this map as a smooth, perfect highway.

However, newer experiments (like Daya Bay) discovered that the real "highway" isn't smooth at all. Around a specific energy level (5 MeV), there is a strange "bump" or a dip in the data that the smooth map didn't predict. It's like driving on a road that suddenly has a pothole or a speed bump that the GPS didn't warn you about.

The Experiment: Redrawing the Map

Guihong Huang asked a simple question: What if we stop using the old, smooth map and instead use the actual, bumpy road measurements from the Daya Bay experiment?

To do this, the author built a new "global analysis framework." Here is how it works, using an analogy:

• The Old Way: Imagine trying to guess the shape of a cake by looking at a drawing of a perfect circle. You assume the cake is perfectly round.
• The New Way: Imagine you have a photo of the actual cake, which has a slightly lopsided frosting and a weird bump on the side. You use that real photo to adjust your guess.

In this study, the author took the raw data from KamLAND (the reactor neutrinos) and combined it with the real, measured spectra from Daya Bay (specifically for Uranium-235 and Plutonium-239). Instead of assuming the neutrinos follow a theoretical curve, the analysis let the real data from Daya Bay "guide" the shape of the curve.

The Results: The Puzzle Pieces Fit Better

When the author swapped the theoretical "smooth map" for the "real, bumpy map," the results changed:

1. The Numbers Shifted: The best-fit values for the "speed of the dance" and the "angle of the dance" moved slightly downward.
2. Better Agreement: These new numbers are now much closer to the measurements from the JUNO experiment.
3. The Tension Relieved: The "tension" (the disagreement) between the old KamLAND results and the new JUNO results became smaller.

The Analogy:
Imagine you are trying to tune a radio to a specific station.

• Scenario A: You use an old, slightly out-of-date frequency guide. You get the station, but there's a lot of static (noise), and the volume is a bit off.
• Scenario B: You update your guide with the actual frequency signal you just measured. Suddenly, the static clears up, and the volume matches perfectly with what your friend (JUNO) is hearing.

The Conclusion

The paper concludes that the disagreement between the KamLAND and JUNO experiments wasn't necessarily because the physics was wrong, but because the theoretical model used to interpret the data was slightly inaccurate.

By using the real-world measurements from Daya Bay to correct the "map," the author showed that the reactor neutrino data actually agrees much better with the solar neutrino data. This suggests that the "bump" in the neutrino spectrum is a real feature of nature that we need to account for to get the most accurate picture of how these particles behave.

In short: The author fixed a "glitch" in the software (the theoretical model) by using real-world data, and suddenly, two different groups of scientists started seeing the same picture.

Technical Summary

Technical Summary: Global Fit of KamLAND Data and the Daya Bay Antineutrino Energy Spectrum

Problem Statement
Recent measurements by the JUNO experiment of solar neutrino oscillation parameters ($\Delta m^2_{21}$ and $\sin^2 \theta_{12}$) show a mild tension (0.2$\sigma$) with the 2013 results from the KamLAND experiment. Concurrently, short-baseline reactor experiments (Daya Bay, RENO, Double Chooz) have identified significant discrepancies between measured antineutrino spectra and the standard Huber-Müller model predictions, specifically a "5 MeV bump" and a total flux deficit. Since reactor neutrino oscillation analyses rely heavily on the assumed antineutrino flux model, it is critical to determine if these spectral anomalies are responsible for the tension between KamLAND and solar neutrino data. Previous global analyses have addressed this but often relied on simplified energy-dependent ratios or lacked a quantitative discussion of the impact on solar parameters.

Methodology
The authors construct a global analysis framework designed to be weakly dependent on the reactor antineutrino flux model. The methodology proceeds in three stages:

1. Reproduction of KamLAND Results: The authors first reproduce the 2013 KamLAND analysis using the Huber-Müller model. They utilize public data regarding reactor thermal power, fission fractions, and detector response. A $\chi^2$ function is constructed based on event counts in 17 energy bins (0.900–8.125 MeV) across three running periods, incorporating systematic uncertainties (energy scale, backgrounds, shape variations) as penalty terms. This step validates the framework, achieving agreement with official KamLAND results within 0.1$\sigma$ when $\theta_{13}$ is constrained.
2. Combined Analysis with Daya Bay Data: The framework is extended to a combined fit of KamLAND data and Daya Bay antineutrino spectra. In this approach:
   • The $^{235}\text{U}$ and $^{239}\text{Pu}$ fission antineutrino spectra are treated as free parameters, binned into 25 energy bins (2–8 MeV).
   • These 50 spectral parameters are constrained by the independently measured, unfolded spectra and covariance matrices from the Daya Bay experiment (2019 publication).
   • The $\chi^2$ function includes a term linking the theoretical expectations in KamLAND to the measured prompt energy spectra of Daya Bay, propagating the spectral constraints to the oscillation parameters.
3. Simplified Comparison: A secondary, simplified analysis is performed using the ratio of the Daya Bay unfolded spectrum to the Huber-Müller prediction (similar to Ref. [10]) to compare against the full combined fit.

Key Contributions

• Framework Development: The paper establishes a global analysis framework that integrates long-baseline (KamLAND) and short-baseline (Daya Bay) data without relying strictly on the Huber-Müller model for the $^{235}\text{U}$ and $^{239}\text{Pu}$ components.
• Quantitative Impact Assessment: It provides a quantitative assessment of how replacing the Huber-Müller model with Daya Bay's measured spectra affects the fitted oscillation parameters.
• Validation: The successful reproduction of KamLAND's 2013 results serves as a rigorous baseline for the subsequent combined analysis.

Results
The combined analysis yields the following shifts in the best-fit values compared to the standard KamLAND-only fit using the Huber-Müller model (with $\theta_{13}$ constrained):

• $\Delta m^2_{21}$: Decreases from $7.53^{+0.17}_{-0.16} \times 10^{-5} \text{ eV}^2$ to $7.50^{+0.19}_{-0.18} \times 10^{-5} \text{ eV}^2$. This represents a shift of approximately 0.18$\sigma$ to 0.19$\sigma$ toward lower values.
• $\tan^2 \theta_{12}$: Decreases from $0.466^{+0.067}_{-0.055}$ to $0.439^{+0.064}_{-0.053}$, a shift of approximately 0.49$\sigma$.
• Comparison with JUNO: These adjusted values show better agreement with the latest JUNO measurements ($\Delta m^2_{21} = 7.50 \times 10^{-5} \text{ eV}^2$, $\tan^2 \theta_{12} = 0.4476$) than the original KamLAND results.
• Method Comparison: The full combined fit (using 2019 isotope-decomposed spectra) produces slightly larger uncertainties than the simplified ratio method, attributed to the inclusion of more independent systematic uncertainties and statistical degrees of freedom. However, the trend of decreasing central values for both parameters remains consistent across both methods.

Significance
The paper claims that the differences in predicted reactor antineutrino spectra (specifically the deviations from the Huber-Müller model observed by Daya Bay) are likely an important cause of the tension between KamLAND and solar neutrino experiments. By demonstrating that incorporating Daya Bay's measured spectra systematically lowers the best-fit values of $\Delta m^2_{21}$ and $\tan^2 \theta_{12}$, the study suggests that reactor neutrino oscillation experiments can effectively break their dependence on specific spectral models. The authors posit that this combined analysis method offers a valuable reference for future systematic uncertainty assessments and spectral model corrections in large-scale neutrino experiments, such as JUNO and Hyper-K.