Test of the GENIE neutrino event generator against reduced cross sections extracted from ${}^{16}$O $(e,e'p)$ data

This paper utilizes precise semiexclusive electron scattering data on oxygen-16 to test and identify deficiencies in the nuclear ground state and final state interaction models within the GENIE neutrino event generator, demonstrating that these models fail to accurately reproduce measured reduced cross sections across all kinematic regions.

The Gist

The Big Picture: Why Do We Need This Test?

Imagine you are trying to solve a mystery: Where did a neutrino come from, and how much energy did it have?

Neutrinos are ghostly particles that zip through the universe. Scientists want to study them to understand the fundamental laws of the universe (like why the universe is made of matter and not just energy). But neutrinos are tricky; they rarely interact with anything. When they do hit an atom, they create a messy explosion of other particles.

To figure out the neutrino's original energy, scientists use a computer program called GENIE. Think of GENIE as a simulation video game. It tries to predict what happens when a neutrino hits an atom. If the game's physics engine is accurate, the scientists can look at the debris (the particles flying out) and work backward to guess the neutrino's speed.

The Problem: The scientists aren't sure if the "physics engine" in GENIE is perfect. If the game is wrong, their calculations about neutrino energy will be wrong, and their whole experiment could be flawed.

The Solution: Borrowing a Blueprint from a Different Game

The authors of this paper had a clever idea. They realized that neutrinos and electrons are like cousins. They both interact with atoms in very similar ways (like two different types of bullets hitting a target).

• Neutrinos are hard to study because we don't know their starting energy, and we can't control the beam perfectly.
• Electrons, however, are easy to control. We can shoot them at atoms with a very precise "gun" and measure exactly what comes out.

The authors decided to use electron data as a "gold standard" blueprint. They took high-precision data from electron experiments (where we know exactly what happened) and used it to test if the GENIE neutrino simulator was playing fair.

The Experiment: The "Oxygen Target"

The scientists focused on Oxygen (specifically the Oxygen-16 isotope). Why? Because many of the world's biggest neutrino detectors (like Super-Kamiokande) are giant tanks of water. Water is made of Hydrogen and Oxygen. So, understanding how neutrinos hit Oxygen is critical for real-world experiments.

They looked at a specific type of collision: Semi-exclusive scattering.

• Analogy: Imagine a billiard table. A cue ball (the lepton) hits a cluster of balls (the nucleus).
• Exclusive: You see exactly which ball flew off and where it went.
• Inclusive: You just see the table shake and count how many balls moved.
• Semi-exclusive: You see the cue ball bounce off and you see exactly one specific ball (a proton) fly out. This gives you a lot of detail about the inside of the cluster.

What They Did

1. The Setup: They took real data from two famous electron experiments (Saclay and NIKHEF) where electrons hit Oxygen and knocked out protons. They calculated a "reduced cross-section."
   • Simple term: Think of this as a "Probability Score." It tells you how likely it is to knock a proton out at a specific speed and angle, after removing all the boring math about the electron beam itself.
2. The Test: They ran the GENIE neutrino simulator (using different settings/models) to see if it could reproduce that same "Probability Score" for Oxygen.
3. The Comparison: They compared the GENIE predictions against the real electron data.

The Results: The Simulator is "Hallucinating"

The results were not good news for the current version of GENIE.

• The Mismatch: The GENIE simulator consistently got the numbers wrong.
  • At low speeds (low momentum), GENIE predicted too many protons would fly out. It was like a video game where the physics engine makes the balls bounce too wildly.
  • At high speeds, GENIE predicted too few protons. It was like the game engine was too "sticky," keeping the balls trapped inside.
• The Cause: The authors found that GENIE's models for the nucleus (the target) were too simple.
  • Analogy: Imagine trying to simulate a crowd of people in a stadium. GENIE treats the crowd like a smooth, uniform fog (a "Fermi Gas"). But in reality, the crowd has structure: people are sitting in specific rows (shells), and they bump into each other in complex ways.
  • GENIE is missing the "texture" of the nucleus. It doesn't account for the specific arrangement of protons and neutrons or how they interact after being hit.

The Takeaway: Why This Matters

The paper concludes that GENIE needs an upgrade.

If neutrino experiments (like those trying to detect dark matter or study neutrino oscillations) rely on a simulator that gets the basic physics of Oxygen wrong, their results will have hidden errors.

The "Silver Lining":
The authors propose a new way to test these simulators. Instead of just guessing, we can use the precise electron data as a ruler to measure the accuracy of neutrino models. It's like using a calibrated scale to check if a bathroom scale is telling the truth.

Summary in One Sentence

The authors used precise data from electron collisions to show that the popular neutrino simulation software (GENIE) is currently using oversimplified models of the atomic nucleus, causing it to predict the wrong number of particles flying out, and they suggest using electron data as a strict "test drive" to fix these models before we rely on them for major neutrino discoveries.

Technical Summary

1. Problem Statement

Current and future accelerator-based neutrino experiments (e.g., DUNE, Hyper-K, T2K) rely heavily on neutrino event generators (such as GENIE) to simulate neutrino-nucleus interactions. These simulations are critical for:

• Reconstructing the incident neutrino energy from final-state particles.
• Estimating backgrounds and systematic uncertainties for neutrino oscillation parameter extraction.

However, the accuracy of these generators is limited by the nuclear models they employ. Specifically, models for the nuclear ground state (momentum distribution, shell structure) and Final State Interactions (FSI) (how the ejected nucleon interacts with the residual nucleus) are often phenomenological or factorized. While these models may describe inclusive cross sections (total rates) reasonably well, they often fail to accurately predict semi-exclusive processes (where specific final-state particles, like a proton, are detected).

The core problem is the lack of rigorous validation of these nuclear models against high-precision data in the kinematic regions relevant to neutrino experiments. Since direct neutrino data with precise kinematics is difficult to obtain, the authors propose using electron scattering data as a proxy, leveraging the similarity between electromagnetic and weak interactions.

2. Methodology

The authors employ a comparative approach using reduced cross sections to test the GENIE event generator (version 3.4.0) against high-precision semi-exclusive electron scattering data on an Oxygen-16 ($^{16}$O) target.

A. Theoretical Framework

• Reduced Cross Section ($\sigma_{red}$): The authors define the reduced cross section by dividing the measured differential cross section by kinematic factors and the elementary lepton-nucleon cross section.
  • $\sigma_{red}$ is identified with the distorted spectral function.
  • Crucially, $\sigma_{red}$ depends primarily on the target's intrinsic properties (nucleon momentum distribution) and FSI, making it largely independent of the interaction type (electromagnetic vs. weak) aside from small Coulomb corrections.
• Relativistic Distorted Wave Impulse Approximation (RDWIA): This microscopic, unfactorized model is used as a benchmark. It successfully describes semi-exclusive electron scattering data and is used to generate "theoretical" neutrino reduced cross sections for comparison.

B. Data Sources

The study utilizes high-statistics, precise kinematics data from two major electron scattering experiments on $^{16}$O:

1. Saclay Experiment: Performed in perpendicular kinematics ($p_m \perp q$).
2. NIKHEF Experiment: Performed in parallel kinematics ($p_m \parallel q$).
   Both datasets cover missing momentum ($p_m$) ranges relevant to quasielastic scattering and missing energy intervals corresponding to the $1p_{1/2}$ and $1p_{3/2}$ shell removals.

C. Simulation Setup

• Target: Oxygen-16 ($^{16}$O), the primary component of water Cherenkov detectors.
• Process: Charged-Current Quasielastic (CCQE) scattering ($\nu_\mu + ^{16}\text{O} \to \mu^- + p + ^{15}\text{N}$).
• GENIE Configurations Tested: Four distinct model sets were simulated:
  1. G18 02a: Relativistic Fermi Gas (RFG) with short-range correlations.
  2. G18 10a: Local Fermi Gas (LFG) with the full Valencia model and RPA.
  3. G21 11 (SuSAv2): Superscaling model using a mean-field model at low $Q^2$ and RFG at high $Q^2$.
  4. effsf: Effective Spectral Function model.
• Event Selection: The simulation selects "1$\mu$1p" events (one muon, one proton, no FSI neutrons) to mimic the $(e, e'p)$ experimental selection.

D. Analysis Strategy

Since GENIE models generally do not resolve specific nuclear shells (like $1s$ vs. $1p$) in the same way electron data does, the authors use a ratio method:

1. Calculate the ratio $R_{1p} = \sigma_{red}(1p) / \sigma_{red}(1p+1s)$ using the RDWIA model.
2. Apply this ratio to the GENIE total reduced cross section to isolate the contribution from the $1p$ shell: $\sigma_{MC}^{red}(1p) = R_{1p} \cdot \sigma_{MC}^{red}(1p+1s)$.
3. Compare this isolated GENIE prediction directly against the experimental electron reduced cross sections.

3. Key Contributions

• Novel Validation Method: The paper establishes a robust method for testing neutrino event generators by mapping their output to electron scattering reduced cross sections. This bypasses the need for direct neutrino data with the same precision.
• Shell-Structure Mapping: The authors developed a technique to extract shell-specific ($1p$ shell) predictions from GENIE models (which typically treat the nucleus as a gas) using RDWIA-derived ratios.
• Systematic Benchmarking: A comprehensive comparison of four major GENIE configurations against two distinct kinematic regimes (perpendicular and parallel) on the same target ($^{16}$O).

4. Results

The comparison revealed persistent and significant disagreements between the GENIE predictions and the experimental electron data across all tested models and kinematic regions.

• Low Missing Momentum ($|p_m| < 100$ MeV/c):
  • Most models (G18 10a, G21 11, effsf) significantly overestimate the reduced cross section.
  • The G18 02a (RFG) model shows agreement here but fails at higher momenta.
• High Missing Momentum ($|p_m| \ge 100$ MeV/c):
  • Models generally underestimate the data (except for G18 02a, which overestimates due to the uniform Fermi gas tail).
  • The G21 11 (SuSAv2) model shows a massive overestimation (up to 300%) at the peak of the distribution in NIKHEF kinematics.
• Kinematic Dependence:
  • Saclay (Perpendicular): GENIE models overestimate at low $p_m$ and underestimate at high $p_m$.
  • NIKHEF (Parallel): The disagreement is even more severe. The GENIE predictions show "absolutely different behavior" compared to the measured data, failing to reproduce the shape of the cross section.
• FSI and Ground State Issues: The discrepancies indicate that the current GENIE models fail to simultaneously describe:
  1. The nuclear ground state (specifically the detailed momentum distribution and shell structure).
  2. The Final State Interactions (attenuation and distortion of the outgoing nucleon).

The RDWIA model, in contrast, showed good agreement with the data, confirming that the issue lies with the phenomenological models in GENIE, not the fundamental physics of the interaction.

5. Significance

• Impact on Oscillation Experiments: Since neutrino energy reconstruction relies on accurate modeling of the hadronic final state (proton kinematics), the failure of GENIE to reproduce semi-exclusive cross sections implies that current systematic uncertainties in oscillation experiments may be underestimated.
• Call for Model Improvement: The paper concludes that neutrino event generators must move beyond simple factorized models (like RFG) and incorporate more sophisticated, microscopic descriptions of nuclear ground states and FSI to accurately simulate CCQE semi-exclusive processes.
• Future Direction: The proposed approach of comparing spectral functions (via reduced cross sections) against precise electron data is highlighted as a "great opportunity" and a "promising method" for validating and refining nuclear physics models in Monte Carlo generators for future neutrino experiments.

In summary, the paper demonstrates that while GENIE is a powerful tool, its current nuclear models are insufficient for precision semi-exclusive physics, necessitating a shift toward more realistic, non-factorized nuclear descriptions to reduce systematic errors in the next generation of neutrino oscillation measurements.