Optimal binning of $D\rightarrow K_{\mathrm S}^0\pi^+\pi^-$ and $D\rightarrow K_{\mathrm S}^0K^+K^-$ phase space for experimental measurements

This paper presents new, optimized binning schemes for the Dalitz plots of $D \rightarrow K_{\mathrm S}^0\pi^+\pi^-$ and $D \rightarrow K_{\mathrm S}^0K^+K^-$ decays, utilizing an improved figure of merit to achieve an estimated 5% gain in the precision of $\gamma$ measurements and a 20% increase in statistical sensitivity for charm mixing observables while accounting for detector resolution and background effects.

The Gist

Imagine you are trying to solve a complex puzzle, but the pieces are scattered across a giant, two-dimensional map. This map represents the "Dalitz plot," a way physicists visualize how particles decay. The goal of this paper is to figure out the best way to draw lines on this map to divide it into sections (bins) so that scientists can extract the most valuable information possible.

Here is a breakdown of what the authors did, using simple analogies:

The Goal: Finding the Angle $\gamma$

Physicists are trying to measure a specific angle in the universe's rulebook, called the CKM angle $\gamma$. Think of this angle as a secret code that explains why the universe is made of matter rather than antimatter. To crack this code, they watch particles called $D$ mesons decay into other particles.

The "map" (Dalitz plot) shows where these decay products land. However, the map is messy. To read the secret code, scientists need to know the "strong phase" (a kind of internal rhythm or timing) of the particles at different spots on the map.

The Old Way vs. The New Way

The Old Way (CLEO_OPTIMAL):
Previously, scientists divided this map into 8 sections based on a simple rule: "Make sure every section has the same amount of 'rhythm change'." It was like cutting a pizza into 8 equal slices. It worked, but it wasn't the most efficient way to find the secret code.

The New Way (NEWGAMMA):
The authors in this paper asked, "Can we cut the pizza differently to get a better taste of the secret code?"

• Better Recipe: They invented a new "scorecard" (a mathematical metric) to judge how good a cut is. Instead of just looking at the rhythm, their new scorecard specifically calculates how much information about the secret angle $\gamma$ is hidden in each slice.
• Accounting for Noise: In the real world, the data isn't clean; there is "background noise" (like static on a radio). The old method ignored this. The new method designs the slices specifically to handle the noise levels found at the LHCb experiment (a giant particle collider). It's like tuning a radio not just to the station, but specifically to the static level in your living room.
• More Slices: They also increased the number of slices from 8 to 10. More slices usually mean more detail, but too many can make the data too sparse to analyze. They found the "Goldilocks" number: 10.

The Result:
By using this new cutting pattern, they estimate they can measure the secret angle $\gamma$ about 5% more precisely than before. It's like upgrading from a standard ruler to a laser measure.

The Second Goal: Studying "Charm Mixing"

There is a second puzzle: studying how these particles "mix" or switch identities over time (called charm mixing).

• The Problem: When you slice the map, particles can sometimes "slip" from one slice into a neighbor due to the blurriness of the detectors (like a ball rolling slightly off a marked line). If you don't account for this, your measurement gets biased (skewed).
• The Solution: For this specific puzzle, the authors created a new cutting pattern called NEWCHARM. They added a "penalty" to their scorecard. If a cut causes too many particles to slip into the wrong slice, the score goes down.
• The Result: This new pattern improves the precision of the mixing measurement by about 20% while keeping the "slippage" error low enough to be ignored.

The Third Puzzle: A Different Particle ($K^0_S K^+ K^-$)

They also looked at a slightly different particle decay ($D \to K^0_S K^+ K^-$). Because this particle is rarer, the map looks different.

• They created three new cutting patterns (with 2, 3, or 4 slices).
• They found that using a 3-slice pattern (OPT_KSKK_3) is the best compromise, offering a 12% improvement in precision over the old 2-slice method.

Why This Matters

Think of the Dalitz plot as a crowded dance floor.

• Old Method: You divide the floor into 8 equal zones and ask people in each zone to shout a number.
• New Method: You realize that the people in the corners are shouting louder and clearer about the secret code, while the people in the middle are harder to hear. So, you draw the zones to capture the loudest, clearest voices, while ignoring the static noise.

Summary of Claims:

1. New Cutting Patterns: They propose new ways to divide the data map for two types of particle decays.
2. Better Math: They used a new formula that specifically targets the precision of the angle $\gamma$ and accounts for background noise.
3. Improved Precision:
   • 5% better precision for measuring the angle $\gamma$.
   • 20% better precision for measuring charm mixing.
4. Safety: They checked that these new patterns don't introduce new errors (like "slippage" or systematic biases) and found them to be safe and robust.

The paper concludes that these new "cuts" are ready to be used by experiments like LHCb and BESIII to get the most accurate measurements possible from their data.

Technical Summary

Technical Summary: Optimal Binning of $D \to K^0_S \pi^+ \pi^-$ and $D \to K^0_S K^+ K^-$ Phase Space

Problem Statement
Measurements of the CKM angle $\gamma$ and charm mixing parameters ($x_{CP}, \Delta x$) rely heavily on the analysis of $D \to K^0_S \pi^+ \pi^-$ and $D \to K^0_S K^+ K^-$ Dalitz plots. These analyses require partitioning the phase space into bins to utilize external strong-phase measurements from quantum-correlated $D^0\bar{D}^0$ pairs (e.g., from BESIII). Current binning schemes, largely based on the CLEO_OPTIMAL design, were optimized using metrics that approximate statistical precision but do not fully capture the sensitivity to $\gamma$ or the specific constraints of charm mixing analyses. Furthermore, existing schemes do not adequately account for modern experimental conditions, such as specific background compositions at LHCb or the migration of events between bins due to detector resolution, which can introduce significant biases in mixing measurements.

Methodology
The authors propose a new framework for determining optimal binning schemes by introducing refined figures of merit (FoM) and incorporating realistic experimental constraints.

• Optimization Metric for $\gamma$: Instead of the traditional metric based on the Fisher information for $x_\pm$ and $y_\pm$ (Eq. 2.7), the authors define a new metric, $Q_\gamma^2$ (Eq. 2.10), based directly on the Fisher information for $\gamma$. This metric utilizes the univariate Fisher information derived from a Poisson likelihood, providing a more accurate proxy for the statistical precision on $\gamma$. For the $D \to K^0_S K^+ K^-$ channel, where correlations between $\gamma$ and the strong phase $\delta_B$ are significant in low-bin-count schemes, a multivariate Fisher information matrix is employed (Eq. 3.5–3.8) to marginalize over nuisance parameters.
• Optimization Metric for Charm Mixing: For the "bin-flip" analysis of charm mixing, the authors define a metric $Q_{x_{CP}}^2$ (Eq. 4.3) based on the sensitivity to $x_{CP}$. Crucially, they introduce a penalty term, $Q_M^2$ (Eq. 4.5), to the metric to minimize the global net migration ($M$) of events between bins caused by detector resolution. The final metric is a weighted sum (Eq. 4.6) balancing statistical sensitivity against migration-induced bias.
• Implementation: The optimization is performed on a $500 \times 500$ grid of sub-bins using a random-walk algorithm. The procedure incorporates:
  • Backgrounds: Realistic background distributions (real-$D$ and combinatorial) observed at LHCb are included in the metric calculation.
  • Efficiencies: Detector efficiency variations are considered, though found to have marginal impact on the optimal scheme.
  • Smoothing: Post-optimization smoothing is applied to remove isolated sub-bin artifacts, with parameters adjusted to match LHCb resolution scales.
• Validation: The performance of new schemes is evaluated using pseudoexperiments that simulate signal yields, background, and detector resolution effects. Systematic uncertainties arising from strong-phase inputs and bin migration are explicitly calculated.

Key Contributions and Results

1. $D \to K^0_S \pi^+ \pi^-$ for $\gamma$ (NEWGAMMA Scheme):

   • A new scheme with $N=10$ bins, named NEWGAMMA, is presented. It is optimized using the new Fisher information metric and includes LHCb background levels.
   • Statistical Gain: Pseudoexperiments estimate a 5% improvement in the precision of $\gamma$ compared to the CLEO_OPTIMAL scheme. The scheme retains 94% of the unbinned statistical sensitivity.
   • Systematics: The propagated systematic uncertainty from strong-phase inputs ($c_i, s_i$) is estimated to be 10% smaller than in CLEO_OPTIMAL. The systematic bias due to bin migration is comparable to that of CLEO_OPTIMAL.
   • Robustness: The scheme remains superior to signal-only optimized schemes across various background scenarios, including LHCb Run 3 conditions.

2. $D \to K^0_S K^+ K^-$ for $\gamma$ (OPT_KSKK Schemes):

   • Three optimized schemes with $N=2, 3, 4$ are presented (OPT_KSKK_2, 3, 4).
   • The OPT_KSKK_3 scheme is highlighted as the optimal compromise, offering a ~12% gain in sensitivity over the current $N=2$ equal-phase scheme while maintaining feasible precision for strong-phase measurements.
   • The optimization accounts for the correlation between $\gamma$ and $\delta_B$, which is critical for low-bin-count schemes.

3. $D \to K^0_S \pi^+ \pi^-$ for Charm Mixing (NEWCHARM Scheme):

   • A dedicated scheme, NEWCHARM ($N=10$), is optimized for the bin-flip analysis of mixing parameters ($x_{CP}, \Delta x$).
   • Bias Mitigation: By including the migration penalty in the optimization, the scheme achieves a statistical sensitivity improvement of ~20% over the current EQUAL_8 scheme while keeping the migration-induced bias at a level comparable to EQUAL_8.
   • Trade-offs: While NEWCHARM improves sensitivity to $x_{CP}$, it increases the uncertainty on $y_{CP}$ by ~35%. The authors note that moving from $N=8$ to $N=10$ yields only marginal gains for mixing sensitivity alone, but the $N=10$ choice is justified by the reduction in strong-phase systematic uncertainties.

Significance and Claims
The paper claims that these new binning schemes represent a necessary evolution for upcoming high-statistics datasets, particularly from BESIII and LHCb. The significance lies in:

• Direct Optimization: Moving away from approximations to metrics that directly target the physics observable of interest ($\gamma$ or $x_{CP}$).
• Realistic Constraints: Explicitly accounting for background compositions and detector resolution effects during the optimization process, rather than treating them as post-analysis corrections.
• Balanced Performance: Achieving significant statistical gains (5% for $\gamma$, 20% for mixing) without introducing prohibitive systematic biases, particularly regarding bin migration.
• Strategic Implementation: The authors propose using NEWGAMMA for all $B^\pm \to DK^\pm$ measurements to ensure consistency in strong-phase inputs, while reserving NEWCHARM specifically for charm mixing analyses where migration bias is a dominant concern. They acknowledge that using two different schemes for the same decay mode prevents a full combination of $\gamma$ and mixing results in the near term but argue this maximizes immediate statistical precision. Future work may allow for a unified scheme if migration corrections become a core part of the analysis framework.