Matter-Induced CPT Violation and Earth-Density Stratification Effects in Long-Baseline Neutrino Oscillation Experiments

Technical Summary: Matter-Induced CPT Violation and Earth-Density Stratification Effects in Long-Baseline Neutrino Oscillation Experiments

Problem Statement
Long-baseline (LBL) neutrino oscillation experiments (T2K, NOνA, DUNE, Hyper-Kamiokande) aim to precisely measure the leptonic CP-violating phase $\delta_{CP}$ and determine the neutrino mass hierarchy. However, these measurements are complicated by two physically coupled systematics arising from the propagation of neutrinos through Earth's matter:

1. Extrinsic CPT Violation: The presence of electrons (but not positrons) in Earth's matter creates a matter potential ($V_f$) that changes sign between neutrinos and antineutrinos. This induces a non-zero asymmetry ($A^{CPT}_{\mu e} \neq 0$) even if CPT symmetry holds exactly in vacuum. This "fake" asymmetry must be subtracted to isolate intrinsic CP violation.
2. Earth Density Stratification: For baselines $L \gtrsim 5000$ km, neutrino trajectories traverse the lower mantle and outer core. The standard approximation of a constant path-averaged density fails to capture these variations, potentially introducing systematic biases in the reconstruction of $\delta_{CP}$.

The core problem addressed is that both effects depend on the same matter-potential Hamiltonian ($H_f \propto \rho(x)$). Consequently, errors in density modeling propagate simultaneously into both the $A^{CPT}_{\mu e}$ estimate and the $\delta_{CP}$ reconstruction. Prior studies have treated these systematics in isolation; this paper argues they must be modeled jointly.

Methodology
The authors employ a unified theoretical and statistical framework:

• Theoretical Framework: The analysis uses exact three-flavor matrix-exponentiation propagators to solve the evolution equation $i d|\nu(x)\rangle/dx = H_f(x)|\nu(x)\rangle$. The Hamiltonian includes the vacuum term (using NuFit 5.3 parameters) and the MSW matter potential derived from the Preliminary Reference Earth Model (PREM).
• Simulation: Neutrino oscillation probabilities are computed for T2K, NOνA, DUNE, and Hyper-Kamiokande. The study evaluates the extrinsic CPT asymmetry defined as $A^{CPT}_{\mu e} = (P_{\mu e} - P_{\bar{e}\bar{\mu}}) / (P_{\mu e} + P_{\bar{e}\bar{\mu}})$.
• Density Uncertainty: The sensitivity of $A^{CPT}_{\mu e}$ to geophysical uncertainties is quantified by applying fractional density shifts $\rho \to \rho(1 + \epsilon)$ where $\epsilon = \pm 1\%, \pm 2\%, \pm 5\%$.
• Statistical Analysis: A Poisson log-likelihood $\chi^2$ statistic is constructed to analyze the bias in $\delta_{CP}$ reconstruction when using a constant-density model against a "truth" generated with the full PREM profile. The analysis includes nuisance parameter pull terms for flux, cross-section, and detector efficiency. Crucially, the authors extend this to an "extended $\chi^2$" that includes a density pull term ($\xi_\rho$) to treat density uncertainty as a correlated systematic affecting both $\delta_{CP}$ and $A^{CPT}_{\mu e}$.

Key Contributions
This work presents one of the first systematic studies to treat matter-induced extrinsic CPT violation and Earth density stratification as correlated systematics within a single statistical framework. Specific contributions include:

1. Joint Treatment: A unified $\chi^2$ framework that simultaneously accounts for density uncertainties affecting both CP-phase reconstruction and CPT asymmetry estimates.
2. Hierarchy Dependence: A quantitative characterization of how the normal (NH) and inverted (IH) mass orderings affect $A^{CPT}_{\mu e}$, revealing a $\sim 9\%$ difference at the DUNE baseline.
3. 3D Surface Analysis: The computation of the $A^{CPT}_{\mu e}(E, \delta_{CP})$ surface, identifying distinct regimes where intrinsic CP and extrinsic CPT effects are entangled (low energy) versus where matter dominates (high energy).
4. Bias Quantification: A rigorous assessment of the reconstruction bias introduced by the constant-density approximation across a wide range of baselines, from current experiments to hypothetical very-long-baseline (VLB) proposals.

Key Results

• Extrinsic CPT Asymmetry: At peak energies, $A^{CPT}_{\mu e}$ ranges from 0.022 (T2K/Hyper-K) to 0.180 (DUNE) under normal hierarchy. The asymmetry is sensitive to $\theta_{23}$, $\theta_{13}$, and $\Delta m^2_{31}$, with a $\sim 9\%$ difference between NH and IH at DUNE.
• Density Sensitivity: A $\pm 5\%$ uncertainty in matter density induces a shift $|\Delta A^{CPT}| \lesssim 0.008$ (approx. 3–4% relative) at DUNE. The ratio definition of $A^{CPT}_{\mu e}$ provides partial self-cancellation of density errors, but the effect remains non-negligible for precision tests.
• 3D Surface Regimes: At DUNE ($L=1285$ km), the $A^{CPT}_{\mu e}$ surface is flat with respect to $\delta_{CP}$ for $E \gtrsim 1.5$ GeV, allowing straightforward subtraction. However, for $E \lesssim 0.5$ GeV, the surface tilts significantly with $\delta_{CP}$ due to the solar oscillation term, requiring joint CP–CPT likelihood analysis to avoid bias.
• Reconstruction Bias:
  • For current experiments ($L \le 1285$ km), the bias in $\delta_{CP}$ reconstruction from using a constant-density model is negligible ($< 0.3^\circ$), well below current experimental precision targets.
  • For proposed VLB facilities ($L > 5000$ km), the bias becomes catastrophic. At $L=7000$ km, the bias is $17.8^\circ$ (3.6 times the DUNE Phase II target); at $L=12000$ km, the bias reaches $172.2^\circ$, effectively reversing the sign of the reconstructed phase.
• Statistical Framework: The extended $\chi^2$ with a density pull term confirms that for current baselines, the constant-density approximation is adequate. However, for VLB proposals, treating density uncertainty as a correlated systematic is mandatory to avoid underestimating total systematics.

Significance and Claims
The paper claims that for the current generation of LBL experiments (T2K, NOνA, DUNE, Hyper-K), the constant-density approximation is sufficient for both CP and CPT analyses, and matter-density uncertainty contributes only a manageable $\sim 1.3\%$ systematic to $A^{CPT}_{\mu e}$.

However, the authors emphasize that for future very-long-baseline proposals (e.g., transcontinental baselines $L > 5000$ km), Earth density stratification and matter-induced CPT violation are not independent refinements but "two faces of the same dominant systematic." The paper concludes that any future analysis of such facilities must utilize the rigorous, correlated statistical framework presented here, employing spatially resolved PREM profiles and joint treatment of density uncertainties to ensure valid CP and CPT measurements.