Worldwide Reactor Neutrino Propagation to Underground Labs: Matter Effects and Flux Predictions

This paper presents a high-precision framework for predicting reactor antineutrino fluxes at underground laboratories by incorporating matter effects (MSW) through both one-dimensional and three-dimensional Earth models, aiming to quantify the impact of Earth's structural features on flux accuracy as geoneutrino measurements approach sub-percent precision.

The Gist

The Big Picture: Listening to the Earth's Heartbeat

Imagine the Earth is a giant, glowing furnace. Deep inside, radioactive elements like Uranium and Thorium are slowly decaying, generating heat that keeps our planet warm and drives volcanoes and earthquakes. Scientists want to "listen" to this heat by catching tiny particles called geoneutrinos that escape from the Earth's core.

However, there's a problem. It's like trying to hear a whisper in a room full of people shouting. The "shouters" are nuclear power plants. They also produce these same particles (antineutrinos) in huge numbers. Because the signals from the Earth and the power plants look almost identical, it's very hard to tell them apart.

The Goal of this Paper:
The authors wanted to build a super-precise "noise-canceling" system. They wanted to calculate exactly how much "shouting" (reactor noise) is coming from every nuclear power plant on Earth to a specific underground lab, so scientists can subtract it and finally hear the Earth's whisper clearly.

---

The Three Main Challenges (and How They Solved Them)

1. The "Global Shouting" Problem

The Issue: There are hundreds of nuclear reactors all over the world. To know the background noise at a lab in China, you need to know the power output of a reactor in France, a plant in the US, and a plant in Japan, all at the exact same time.
The Solution: The team acted like a global data detective. They scraped data from the International Atomic Energy Agency (IAEA) to build a massive, up-to-date list of every operating reactor, how much power they are making, and exactly where they are located. They then calculated the distance from every single one of these reactors to major underground labs (like the one in China's JinPing mountain).

2. The "Traveling Through Rock" Problem (Matter Effects)

The Issue: This is the most scientific part. When neutrinos travel through empty space, they behave one way. But when they travel through the Earth, they have to squeeze through layers of rock, the mantle, and the core.

• The Analogy: Imagine a group of runners (neutrinos) running a race.
  • In a vacuum (space): They run on a flat, empty track. They stay in their lanes perfectly.
  • Through the Earth: The track changes. Sometimes it's mud (the crust), sometimes it's a dense forest (the mantle), and sometimes it's a thick swamp (the core). As they run through these different terrains, the runners start to trip, swap lanes, or change their speed slightly.
• The Science: This is called the MSW effect. For a long time, scientists thought this "swamp effect" was so small it didn't matter. But now, we are trying to measure things with such high precision (better than 1%) that even a tiny change in the runners' path matters. If you ignore the mud, your prediction of who wins the race will be slightly wrong.

The Solution: The authors built a sophisticated computer simulation. Instead of just guessing, they used a mathematical trick called Strang-splitting.

• The Analogy: Imagine you are walking through a city with changing traffic lights. Instead of trying to calculate the whole trip in one giant, messy math equation, you break the trip into tiny steps. At each step, you calculate the traffic (rock density) and the walking speed (neutrino energy) separately, then combine them. This allows them to calculate the path through the Earth's complex layers with extreme speed and accuracy.

3. The "3D Map" vs. "Flat Map" Problem

The Issue: Most old models treated the Earth like a perfect, layered onion (1D model). But the Earth isn't a perfect onion; it has mountains, trenches, and weird blobs of dense rock in the mantle (3D model).
The Solution: They created a hybrid map. For the deep core, they used the standard "onion" model. But for the top 300km (where the crust and upper mantle are), they used a detailed 3D map that accounts for the actual lumpy, uneven density of the Earth's surface.

---

What Did They Find?

After running their high-precision simulations, here is what they discovered:

1. The "Mud" Matters: Even though the Earth's rock only changes the neutrino path by a tiny fraction (less than 1%), this is huge for modern experiments. It's like trying to measure the weight of a feather with a scale that is accurate to the microgram; if you ignore the air pressure, your measurement is wrong.
2. Location is Key:
   • JUNO (China) and Yemilab (South Korea): These labs are surrounded by many medium-distance reactors. Because the neutrinos travel a "medium" distance through the Earth, the "mud effect" is strongest here. The difference between ignoring the Earth's rock and including it is about 0.5%.
   • Boulby (UK): This lab is very close to a few big reactors. Because the distance is short, the neutrinos don't have enough time to get "tangled" in the Earth's rock. The effect is tiny here.
3. The Result: By including these "Earth-rock" corrections, the team can predict the background noise from nuclear plants with much higher accuracy.

Why Does This Matter?

Think of it like tuning a radio.

• Before: Scientists were trying to tune into the "Earth Heat" station, but the "Nuclear Plant" station was bleeding into the signal, making it fuzzy.
• Now: This paper provides a crystal-clear map of exactly how loud the "Nuclear Plant" station is at every frequency.

With this new, ultra-precise calculation, scientists can finally subtract the reactor noise with sub-percent accuracy. This will allow them to finally measure exactly how much heat the Earth generates from radioactive decay. This helps us answer big questions: How much of the Earth's heat comes from radioactivity? How much comes from leftover heat from the planet's formation?

In short: They built a better calculator to help us hear the Earth's heartbeat by perfectly silencing the noise of human industry.

Technical Summary

1. Problem Statement

Geoneutrinos (electron antineutrinos, $\bar{\nu}_e$, from the decay of $^{238}\text{U}$, $^{232}\text{Th}$, and $^{40}\text{K}$) are crucial for understanding Earth's internal heat sources and composition. However, their detection is severely complicated by a dominant background: electron antineutrinos produced by commercial nuclear reactors.

• The Challenge: As geoneutrino measurements and reactor background modeling approach sub-percent precision, previously negligible effects become significant. Specifically, the Mikheyev–Smirnov–Wolfenstein (MSW) matter effect—where neutrinos interact with electrons in the Earth's mantle and core—induces small corrections to reactor antineutrino oscillation probabilities.
• The Gap: While these corrections are typically $<1\%$ in the MeV energy range, they are now comparable to the target precision of future experiments (e.g., JUNO, CJPL, SNO+). Existing models often neglect these effects or rely on computationally expensive methods (repeated full Hamiltonian diagonalization) that are impractical for global-scale Monte Carlo simulations involving thousands of reactor-detector trajectories.

2. Methodology

The authors developed a high-precision prediction framework to quantify reactor $\bar{\nu}_e$ fluxes at underground laboratories, incorporating global data and advanced propagation physics.

A. Data Compilation

• Reactor Fleet: Utilized the IAEA Power Reactor Information System (PRIS) and INFN Ferrara database to compile data on ~413 operating reactors (2024 baseline) and projected configurations for 2026 and 2030.
• Flux Calculation: Calculated thermal power ($P_{th}$) from electrical power ($P_e$) using reactor-type specific thermal efficiencies and load factors (LF).
• Spectral Model: Adopted the SM2023 Summation Model for isotope-dependent spectra ($^{235}\text{U}, ^{238}\text{U}, ^{239}\text{Pu}, ^{241}\text{Pu}$) and used PDG 2025 oscillation parameters.

B. Propagation Solvers

To solve the three-flavor neutrino evolution equation through the Earth, the authors implemented two approaches:

1. Second-Order Strang-Splitting Solver (Primary Method):
   • Implemented in the vacuum mass basis.
   • Decomposes the evolution operator for each layer step: $e^{-i(H_{vac} + H_{mat})\Delta x} \approx e^{-iH_{vac}\Delta x/2} e^{-iH_{mat}\Delta x} e^{-iH_{vac}\Delta x/2}$.
   • Advantage: Avoids repeated diagonalization of the full Hamiltonian. The vacuum term is diagonal, and the matter term is a rank-1 projector, making the calculation computationally efficient and easily parallelizable.
   • Accuracy: Second-order global accuracy with local error $O(\Delta x^3)$.
2. First-Order Analytical Approximation:
   • Derived as a complementary diagnostic tool based on layer-to-layer rotation expansions (SU(3) and SU(2) limits).
   • Used to benchmark the numerical solver.

C. Earth Density Models

The study evaluated fluxes using three distinct Earth models:

1. Vacuum: No matter effects.
2. 1D Model: Preliminary Reference Earth Model (PREM) with spherically symmetric density.
3. Hybrid 3D Model:
   • 0–320 km: LITHO1.0 (crust/upper mantle) with explicit density and material-dependent electron fractions.
   • 320–2880 km: SAW642AN (mantle) using shear-velocity scaling for density.
   • >2880 km: PREM core profile.

D. Uncertainty Propagation

A factorized Monte Carlo strategy was employed:

• Reactor Sector: Propagated uncertainties in fission fractions, thermal power, load factors, and SM2023 spectral shapes ($3 \times 10^5$ realizations).
• Oscillation Sector: Propagated oscillation parameter uncertainties ($1500$ MSW samples, $10000$ vacuum samples).
• Systematics: Defined specific metrics for solver dependence ($\Delta_M$), Earth model dependence ($\Delta_\oplus$), and mass ordering dependence ($\Delta_{N/I}$).

3. Key Contributions

• Efficient Solver: Introduced a Strang-splitting algorithm that enables high-precision MSW calculations for global reactor fleets without the prohibitive cost of full Hamiltonian diagonalization at every step.
• 3D Earth Modeling: Integrated lateral density inhomogeneities (via LITHO1.0 and SAW642AN) into reactor neutrino flux predictions, moving beyond simple 1D spherical approximations.
• Quantitative Assessment: Provided the first systematic quantification of MSW corrections and Earth-model dependencies for a global network of underground labs (including CJPL, JUNO, Yemilab, Boulby, etc.) for current and future years (2026, 2030).

4. Key Results

• Magnitude of MSW Effect:
  • For CJPL (China JinPing), the inclusion of MSW effects increases the integrated reactor flux by ~0.3% and the Inverse Beta Decay (IBD) event rate by ~0.7% compared to vacuum predictions.
  • While small, these corrections are significant for sub-percent precision goals.
• Earth Model Dependence ($\Delta_\oplus$):
  • The difference between 1D (PREM) and 3D Earth models ranges from 0.05% to 0.53% across different sites.
  • JUNO and Yemilab show the largest sensitivity (0.5%) because their backgrounds are dominated by medium-baseline reactors (50–70 km) where matter effects accumulate significantly.
  • Boulby shows the smallest sensitivity (0.05%) because its dominant background comes from very short baselines (24 km) where matter effects are negligible.
• Solver Convergence: The Strang-splitting method (with $N_x=8000$ layers) converges to the reference solution with a relative deviation of $<10^{-6}$ for IBD rates, validating its use as the baseline method. The first-order approximation is adequate for diagnostics but has a systematic offset of $\sim 10^{-4}$.
• Future Projections: Predictions for 2026 and 2030 show moderate changes in flux, with specific sites like CJPL and Kamioka seeing increases due to new reactor operations, while SURF sees a slight decrease due to the Diablo Canyon shutdown.

5. Significance

• Precision Geoneutrino Science: This work establishes that Earth-matter effects and 3D density variations are no longer negligible for reactor background modeling. Ignoring them introduces systematic errors that exceed the statistical precision of next-generation experiments.
• Background Suppression: Accurate modeling of the reactor background is essential for isolating the geoneutrino signal. By quantifying the MSW correction and Earth-model dependence, this framework allows experiments to subtract the reactor background with higher fidelity, improving constraints on Earth's radiogenic heat and U/Th ratios.
• Methodological Standard: The proposed Strang-splitting solver provides a computationally efficient standard for future global neutrino propagation studies, enabling the inclusion of complex Earth models in large-scale Monte Carlo simulations.

In conclusion, the paper demonstrates that to achieve the sub-percent precision required for future geoneutrino experiments, reactor background predictions must incorporate high-fidelity MSW propagation through realistic 3D Earth models.