Correlated Matter Induced Biases in Long-Baseline Neutrino Oscillation Measurements

This paper demonstrates that approximating Earth's matter density as constant in long-baseline neutrino oscillation analyses introduces fundamental systematic errors that propagate across all channels via PMNS unitarity, with the $\nu_{\mu}\rightarrow\nu_{\tau}$ channel exhibiting the largest biases and necessitating spatially resolved density treatments for future precision experiments.

The Gist

Imagine you are trying to listen to a specific radio station (neutrinos) while driving a car (the Earth) through a landscape that changes constantly. To understand the music clearly, you need to know exactly how the terrain affects the signal.

This paper argues that scientists have been using a "flat map" to navigate a "hilly world," and this mistake is causing them to mishear the music, especially on very long trips.

Here is a breakdown of the paper's findings using simple analogies:

1. The Problem: The "Flat Map" Mistake

Scientists study neutrinos (ghostly particles) to learn secrets about the universe, specifically a property called CP violation (which helps explain why the universe is made of matter and not just energy). To do this, they shoot neutrinos from a source, through the Earth, to a detector thousands of kilometers away.

As these particles travel, they interact with the electrons in the Earth's rocks. This interaction changes how the particles "oscillate" (switch flavors).

• The Old Way: Scientists have been treating the Earth like a giant, uniform block of cheese. They assume the density (how packed the rocks are) is the same everywhere along the path. They take an average and use that single number for the whole trip.
• The Reality: The Earth is more like a layered cake with different densities in the crust, mantle, and core, and it has "bumps" and "valleys" (geological fluctuations) that aren't perfectly smooth.

The paper says that using the "flat map" (constant density) instead of the "real terrain" (PREM profile) introduces a systematic error. It's not just a small typo; it's a fundamental misunderstanding of the path.

2. The Domino Effect: The "Three-Flavor Balancing Act"

Neutrinos come in three flavors: electron, muon, and tau. The laws of physics (specifically a rule called unitarity) say that the total probability of these flavors must always add up to 100%. Think of it like a three-legged stool or a balanced scale.

• The Paper's Discovery: If you get the density wrong, you don't just mess up the measurement for one flavor. Because the flavors are mathematically linked, a mistake in the electron channel forces a compensating, correlated mistake in the muon and tau channels.
• The Analogy: Imagine a seesaw with three kids. If you push down on the left side (the electron channel), the other two sides (muon and tau) must go up to keep it balanced. You can't just fix the left side without realizing the other two are now tilted in a specific, predictable way. The paper shows that the "tau" channel is actually the most sensitive and volatile part of this seesaw, carrying the biggest "wobble" caused by the bad map.

3. The Distance Matters: Short vs. Long Trips

The paper tested this at different distances (baselines):

• Short Trips (Under 4,000 km): Like driving across a small town. The terrain is relatively flat and uniform. Using a "flat map" here works fine. The error is tiny (less than 1 degree of error in the measurement).
• Long Trips (Over 5,000 km): Like driving across a continent, going deep into the Earth's mantle and core. Here, the density changes drastically.
  • The Result: Once you pass the 5,000 km mark, the "flat map" assumption breaks down completely. The error explodes.
  • The Consequence: At 12,000 km, the error becomes so huge (over 100 degrees) that the measurement becomes useless. It's like trying to navigate a trans-Atlantic flight using a map of your local neighborhood; you will end up in the wrong ocean.

4. Why Adding More Data Doesn't Help

Usually, in science, if you have more data or look at more channels, you can cancel out errors.

• The Surprise: The paper found that because the errors in the three channels are locked together by the laws of physics, adding more data doesn't fix the problem.
• The Analogy: Imagine you are trying to find the true weight of an object, but your scale is broken in a way that makes it read 10% heavy for every object you weigh. If you weigh the object three times, you don't get the average; you just get a very confident, very wrong answer three times. The "joint fit" (combining all channels) actually reinforces the wrong answer because the error is consistent across the board.

5. The Bottom Line

The authors conclude that for future, ultra-precise experiments (especially those looking at very long distances or combining data from different sources), we cannot treat the Earth as a simple, average block.

• The Takeaway: To get the correct answer about the universe's secrets, scientists must use spatially resolved density treatments. They need to account for the actual, bumpy, layered structure of the Earth, not just an average.
• The Limit: There is a "geophysical sensitivity floor." If you try to measure these particles with extreme precision over long distances without modeling the Earth's real density, you will hit a wall of error that no amount of better detectors can fix. The Earth's geology itself becomes the limiting factor in the measurement.

In short: You can't measure the universe's secrets accurately if you don't accurately model the dirt you're shooting through.

Technical Summary

Technical Summary: Correlated Matter Induced Biases in Long-Baseline Neutrino Oscillation Measurements

Problem Statement
The determination of fundamental neutrino parameters, specifically the leptonic CP-violating phase ($\delta_{CP}$) and the $\theta_{23}$ octant, is entering a hyper-precision era. However, as statistical uncertainties diminish in next-generation long-baseline (LBL) experiments like DUNE and Hyper-Kamiokande, sensitivity becomes increasingly limited by systematic environmental effects. The primary challenge is the Mikheyev-Smirnov-Wolfenstein (MSW) effect, where coherent forward scattering of neutrinos off ambient electrons introduces an extrinsic matter potential ($V_{eff}$) that competes with intrinsic CP violation.

Standard experimental analyses typically mitigate this by employing simplified path-averaged constant-density profiles or globally scaled variations of the Preliminary Reference Earth Model (PREM). This paper challenges that paradigm, arguing that spatially correlated variations in Earth's matter density profile break these simplifications. The authors posit that highly localized, correlated fluctuations along the baseline introduce energy-dependent phase shifts that cannot be captured by a single marginalized effective density parameter. If uncorrected, these geophysical uncertainties project directly into the $(\theta_{23}, \delta_{CP})$ parameter space, inducing structural biases and spurious degenerate solutions.

Methodology
The study employs a rigorous numerical framework to evaluate these effects without relying on idealized profiles:

• Exact Numerical Propagation: The authors solve the three-flavor Schrödinger equation via exact numerical exponentiation, avoiding adiabatic assumptions and retaining full non-commutative evolution effects. This contrasts with analytic treatments (e.g., Cervera-Freund approximation) that assume constant or slowly varying matter potentials.
• Realistic Density Models: Synthetic event spectra are generated using the multi-layer PREM profile (representing mantle-core-mantle layering) as the "true" data. These are subsequently fitted assuming a path-averaged constant density ($\rho_0$).
• Stochastic Analysis: To model geophysical uncertainties, the study generates spatially correlated density fluctuations using Cholesky decomposition with varying correlation lengths ($\ell_c$). These fluctuations are modeled as Gaussian random fields with a Matérn-like exponential covariance, simulating localized geological features like crustal discontinuities.
• Unitarity-Residual Analysis: The study evaluates the probability difference $\Delta P_{\mu\alpha} = P^{\text{PREM}}_{\mu\alpha} - P^{\text{const}}_{\mu\alpha}$ across channels. It enforces the constraint that the sum of these deviations must vanish ($\sum_\alpha \Delta P_{\mu\alpha} = 0$) due to PMNS unitarity.
• Statistical Framework: A Poisson log-likelihood ($\chi^2$) minimization is used to extract the induced bias in $\delta_{CP}$ over baselines ranging from 1,000 to 12,000 km. The analysis compares reconstructed $\delta_{CP}$ against an injected truth of $-90^\circ$.

Key Contributions and Results

1. Correlated Multi-Channel Bias: The paper demonstrates that matter-profile mismodeling does not merely affect the standard $\nu_\mu \to \nu_e$ appearance channel. Due to PMNS unitarity, any density-induced bias in $P_{\mu e}$ is algebraically anti-correlated with compensating shifts in $P_{\mu\tau}$ and $P_{\mu\mu}$. Consequently, the systematic error is inherently coupled across the full oscillation probability space.
2. The $\nu_\mu \to \nu_\tau$ Channel as a Volatile Carrier: A critical finding is that the $\nu_\mu \to \nu_\tau$ channel is the most volatile carrier of the geophysical systematic. Across varying correlation lengths at baselines of 5,000 km and 7,000 km, the $\tau$-appearance channel consistently exhibits a larger mean bias and variance than the standard $\nu_\mu \to \nu_e$ channel.
3. Baseline Dependence and Thresholds:
   • Short Baselines ($L \lesssim 4,000$ km): The bias remains negligible ($|\Delta \delta_{CP}| < 1^\circ$). At DUNE-like distances (1,300 km), the matter column is sufficiently homogeneous that constant-density approximations are adequate.
   • Intermediate Baselines ($L \sim 5,000$ km): A sharp onset of significant bias occurs as neutrino paths begin traversing denser mantle and core-adjacent layers.
   • Long Baselines ($L \ge 5,000$ km): The bias becomes experimentally disqualifying, exceeding $5^\circ$–$10^\circ$. At $L = 12,000$ km, the bias diverges catastrophically (reaching $\sim 177^\circ$), rendering reliable $\delta_{CP}$ reconstruction impossible without accurate Earth matter modeling.
4. Failure of Joint Fits: The study shows that a joint fit combining all oscillation channels fails to mitigate the density-induced bias. Because the constant-density approximation introduces coherent errors across all channels that point toward the same erroneous $\delta_{CP}$ minimum, adding channels reinforces the incorrect best-fit rather than averaging it away.
5. Nature of the Systematic: The bias is identified as a long-wavelength, baseline-integrated effect driven by the total matter column. It is not a short-wavelength noise effect that can be suppressed by smoothing or averaging the density model; rather, it persists across a wide range of correlation lengths ($\ell_c$).

Significance
The paper concludes that spatially resolved density treatments are a mathematical necessity for the analysis frameworks of future precision neutrino facilities. The findings establish a "geophysical sensitivity floor" for next-generation measurements. While individual standalone baselines (like DUNE or Hyper-K) may remain resilient, future global joint analyses that combine atmospheric neutrino event rates (which span core-crossing trajectories up to 12,000 km) with long-baseline accelerator data will immediately encounter these non-linear, correlated matter biases. The work defines the boundary (approximately 5,000 km) where uniform-density frameworks catastrophically break down, necessitating a shift toward stochastic, spatially resolved matter modeling to preserve the physical validity of future CP violation and mass-ordering measurements.