⚛️ phenomenology

Testing the Gaussian Ansatz for TMD PDFs Through CMS Measurements

This paper compares the Collins-Soper-Sterman evolution method and the Gaussian Ansatz for TMD PDFs against CMS Drell-Yan data at 13 TeV, finding that while the CSS method accurately describes the data across all energy ranges, the Gaussian Ansatz is limited to low-energy regions.

Original authors: Shuailiang Yang, Yingqi Hou, Xiaoyu Wang

Published 2026-10-28

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Shuailiang Yang, Yingqi Hou, Xiaoyu Wang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Inside every atom, protons are not solid, featureless spheres but bustling cities of smaller particles called quarks and gluons. For decades, physicists have mapped how these particles move forward and backward along the length of the proton, but a complete picture requires understanding how they move sideways, too. This sideways motion is a key to unlocking the three-dimensional structure of matter, yet it is governed by the complex, invisible forces of the strong interaction that bind the universe together. To see this hidden landscape, scientists look at high-energy collisions where protons smash together, creating pairs of leptons that fly apart. By measuring the sideways speed of these particles, researchers can infer the internal dynamics of the proton. However, the mathematics required to describe this motion changes depending on the energy of the collision, and for a long time, physicists relied on a simplified model that worked well at low energies but might fail when the stakes get higher.

A team of researchers set out to test the limits of this simplified model against a more rigorous, evolving framework. They focused on a specific process known as the Drell-Yan reaction, which occurs when two protons collide and produce a pair of leptons. Using data collected by the CMS experiment at the Large Hadron Collider, where protons collide at an energy of 13 trillion electron volts, the team analyzed the distribution of these lepton pairs. The data covered a vast range of energies, from the mass of the Z boson up to much higher values, allowing the scientists to see how the sideways motion of the particles behaved across different scales. They compared two different ways of calculating the theoretical predictions: one method, known as the Gaussian Ansatz, assumes the sideways motion follows a simple, bell-shaped curve that stays the same regardless of energy. The other method, the Collins-Soper-Sterman evolution, treats the sideways motion as a dynamic quantity that changes and evolves as the energy of the collision increases, accounting for the complex interactions of quantum chromodynamics.

The results of this comparison were clear and decisive. When the researchers applied the simple, bell-shaped curve model to the high-energy data, it failed to match what was actually observed. The model worked reasonably well for the lower energy collisions, but as the energy increased, its predictions diverged sharply from the experimental measurements, underestimating the spread of the particles. In contrast, the more complex evolution method, which accounts for how the internal structure of the proton changes with energy, aligned perfectly with the data across the entire range. The theoretical curves generated by this advanced method traced the path of the experimental points with high precision, capturing the subtle shifts in the particle distribution that the simple model missed. This agreement held true for all the different energy bins the team examined, from the lowest masses up to the highest energies recorded in the dataset.

This study confirms that the internal world of the proton is far more dynamic than a static snapshot suggests. The simple assumption that the sideways motion of quarks remains fixed is insufficient for describing the high-energy environment of modern particle colliders. Instead, the behavior of these particles is deeply tied to the energy scale of the interaction, evolving in a way that only a sophisticated theoretical framework can describe. By validating the more complex evolution method against real-world data, the researchers have reinforced the reliability of the tools used to probe the three-dimensional structure of matter. Their work demonstrates that while simplified models have their place in lower-energy regimes, a full understanding of the nucleon's internal architecture requires a theory that respects the changing nature of the strong force at high energies. This finding provides a solid foundation for future experiments, ensuring that the maps physicists draw of the proton's interior are built on a description of reality that holds up under the most extreme conditions.

Technical Summary: Testing the Gaussian Ansatz for TMD PDFs Through CMS Measurements

Problem Statement
Transverse Momentum Dependent Parton Distribution Functions (TMD PDFs) are essential for mapping the three-dimensional internal structure of nucleons, encoding non-perturbative QCD information regarding parton transverse motion ( $k_T$ ). A central challenge in TMD phenomenology is determining the validity of different modeling approaches for the transverse momentum dependence of these functions. Historically, the "Gaussian Ansatz" (Generalized Parton Model), which assumes a fixed, factorized Gaussian form for transverse momentum, has successfully described low-energy data. However, modern TMD factorization theorems, specifically the Collins-Soper-Sterman (CSS) formalism, predict that transverse momentum distributions evolve with the hard energy scale ( $Q$ ) via a Sudakov-like form factor, deviating from a fixed Gaussian shape at high energies. This study investigates whether the simplified Gaussian Ansatz remains valid at the high-energy scales probed by the Large Hadron Collider (LHC) or if the full CSS evolution is required to describe experimental data.

Methodology
The authors perform a phenomenological analysis of the differential cross section for the proton-proton Drell-Yan process ( $pp \to \gamma^*/Z \to \ell^+\ell^- + X$ ) within the framework of TMD factorization. The study compares two distinct theoretical approaches against experimental data from the CMS Collaboration at $\sqrt{s} = 13$ TeV (integrated luminosity of $36.3 \text{ fb}^{-1}$ ):

CSS Evolution Method: This approach utilizes the rigorous CSS formalism. The calculation is performed in impact parameter ( $b$ ) space to handle the resummation of large logarithmic terms. Key components include:
- Evolution: The evolution from an initial scale $\mu_B$ to the hard scale $Q$ is encoded in a Sudakov form factor, $S(Q, b)$ , which separates into perturbative ( $S_{pert}$ ) and non-perturbative ( $S_{NP}$ ) parts.
- Matching: A $b_*$ prescription ( $b_* = b/\sqrt{1+b^2/b_{max}^2}$ ) is employed to match perturbative calculations at small $b$ with non-perturbative parameters extracted from data at large $b$ .
- Hard Factors: The calculation incorporates scheme-dependent hard factors (Collins-11 and Ji-Ma-Yuan schemes) and Wilson coefficients ( $C$ -coefficients) to ensure consistency with the renormalization group equations.
- Electroweak Corrections: For invariant masses near and above the $Z$ -boson mass, the model includes electroweak couplings and $Z$ -boson contributions.
Gaussian Ansatz: This approach assumes a simplified, scale-independent transverse momentum dependence where the TMD PDFs follow a fixed Gaussian distribution characterized by a mean squared transverse momentum $\langle k_T^2 \rangle = 0.25 \text{ GeV}^2$ . This method neglects the QCD evolution of the transverse momentum distribution.

The differential cross sections ( $d\sigma/dq_T$ ) calculated via both methods are integrated over the kinematic ranges of the CMS data (rapidity $y \in [-2.4, 2.4]$ and specific dilepton invariant mass bins: $50 < m_{ll} < 76$ GeV up to $350 < m_{ll} < 1000$ GeV) and compared directly with the measured data.

Key Contributions and Results

Validation of CSS Evolution: The results demonstrate that the CSS evolution method provides a consistent and accurate description of the differential cross section across the entire energy range tested. The theoretical predictions align well with the CMS data for all dilepton invariant mass bins, from the low-mass region ($50-76$ GeV) to the high-mass region ($350-1000$ GeV).
Limitations of the Gaussian Ansatz: In contrast, the Gaussian Ansatz shows significant deviations from experimental data as the energy scale increases. While the Gaussian model agrees with the CSS predictions and experimental data in the low-energy region ( $Q \sim 1-10$ GeV), it fails to describe the data at higher energies ( $Q \geq 50$ GeV). Specifically, the Gaussian model predicts a rapid decrease in the structure function that does not match the observed $q_T$ spectrum at high $Q$ .
Quantitative Comparison: Numerical estimates of the structure functions at fixed momentum fractions ( $x_1=0.1, x_2=0.2$ ) reveal that the Gaussian method diverges from the CSS results as $Q$ increases from 1 GeV to 200 GeV, confirming that the intrinsic transverse momentum distribution is not static but evolves with the hard scale.

Significance
The study verifies the effectiveness of the TMD factorization theorem, specifically the CSS evolution formalism, in describing high-energy collider processes. It establishes that while the Gaussian Ansatz is a viable approximation for low-energy phenomenology, it is insufficient for high-energy regimes where QCD evolution effects become dominant. These findings provide theoretical support for using the full CSS evolution framework to explore the nucleon's three-dimensional partonic structure and non-perturbative QCD dynamics in future high-energy experiments. The work underscores the necessity of accounting for scale-dependent evolution when interpreting TMD observables at the LHC.

More like this