Effective DPS Cross Section

• Effective double-parton scattering cross section is a key parameter that quantifies the transverse overlap area for independent parton scatterings in high-energy collisions.
• Experimental strategies isolate DPS signals using template fits and multivariate analyses to extract σ_eff with careful control of systematic uncertainties.
• Theoretical models including Light-Front, holographic QCD, and Gaussian profiles reveal explicit x and process dependence, deepening our understanding of nucleon structure.

The effective double-parton-scattering (DPS) cross section, denoted $\sigma_\mathrm{eff}$, is a central phenomenological parameter in the study of multiparton interactions at high-energy colliders. It quantifies the effective transverse area for independent partonic scatterings in proton–proton (pp), proton–nucleus (pA), or nucleus–nucleus (AA) collisions, and encapsulates the impact of both spatial and dynamical correlations among partons. Precise definitions, experimental extraction methodologies, and theoretical interpretations of $\sigma_\mathrm{eff}$ have evolved considerably, as detailed below.

1. Definition and Theoretical Framework

In the factorized approximation, the inclusive DPS cross section for producing final states $A$ and $B$ in the same hadronic collision is commonly expressed as

$$\sigma^{\mathrm{DPS}}_{A+B} = \frac{m}{2}\, \frac{\sigma_A\,\sigma_B}{\sigma_\mathrm{eff}},$$

where $\sigma_A$ and $\sigma_B$ are the single-parton-scattering (SPS) cross sections for $A$ and $B$ respectively, and $m$ is a symmetry factor ($m=2$ for $A\neq B$, $m=1$ for $A=B$). The effective cross section $\sigma_\mathrm{eff}$ parameterizes the inverse of the typical transverse overlap of partons in the proton and has dimensions of area, typically quoted in millibarns (mb) (Collaboration, 2013, d'Enterria et al., 2017).

In more general terms, incorporating the structure of the proton through double parton distributions (DPDs), the DPS cross section is a convolution:

$$\sigma^{\mathrm{DPS}}_{A+B} = \frac{m}{2}\sum_{i,j;k',l'}\int dx_1 dx_2 dx_1' dx_2'\, d^2\mathbf{r}\;\Gamma_{ij}^p(x_1,x_2;\mathbf{r})\,\hat{\sigma}_{ik'}^A(x_1,x_1')\,\hat{\sigma}_{jl'}^B(x_2,x_2')\,\Gamma_{k'l'}^p(x_1',x_2';\mathbf{r}),$$

where $\Gamma_{ij}(x_1, x_2; \mathbf{r})$ denotes the DPDs and $\mathbf{r}$ is the transverse separation between partons. Neglecting nontrivial correlations and integrating over kinematics allows one to identify an experimentally extractable, process- and kinematics-dependent $\sigma_\mathrm{eff}$ (Lovato et al., 5 Jun 2025, Ostapchenko et al., 2015).

2. Experimental Determination

The extraction of $\sigma_\mathrm{eff}$ from data relies on identifying and isolating the DPS contribution in selected final states. This is achieved through:

• Template or Likelihood Fits: Experimental signatures of DPS (such as decorrelated topologies and broader angular distributions) are modeled via mixed or overlaid single-scattering events, while SPS backgrounds are modeled with dedicated Monte Carlo samples that exclude multiple hard interactions (Collaboration, 2013, Sadeh, 2013, Collaboration, 2021).
• Key Observables:
  • $$\Delta^{\mathrm{rel}} p_T = \frac{|\mathbf{p}_{T,1} + \mathbf{p}_{T,2}|}{|\mathbf{p}_{T,1}| + |\mathbf{p}_{T,2}|}$$, measuring the vector balance of two subsystems.
  • $\Delta S$, the azimuthal angle between subsystems.
  • Specialized discriminant variables in four-jet and $W + 2$ jet topologies (Collaboration, 2013, Sadeh, 2013).
  • Multivariate analyses (e.g., neural network outputs or Boosted Decision Trees) in complex channels (e.g., same-sign WW) (Collaboration, 13 May 2025).
• Cross Section Extraction: Once the DPS-enriched fraction $f_\textrm{DPS}$ is measured, the effective cross section is calculated using formulae of the form

$$\sigma_\mathrm{eff} = \frac{R}{f_\mathrm{DPS}\,\sigma^\prime_{2j}}$$

(for $R$ and $\sigma^\prime_{2j}$ as defined in (Collaboration, 2013)) or analogous expressions calibrated for the experimental setup.

Systematic uncertainties arise from template model dependence, jet and lepton energy calibrations, unfolding methodologies, and background subtractions. Modern analyses often quote both statistical and systematic errors explicitly (Collaboration, 2013, Sadeh, 2013, Collaboration, 2021).

3. Universality and Process Dependence

Traditionally, $\sigma_\mathrm{eff}$ was hypothesized to be universal, reflecting the geometric area of parton overlap in the proton. However, extensive analyses across a variety of final states (jets, electroweak bosons, heavy quarkonia) and collision energies have demonstrated significant variation, contradicting universality (Lovato et al., 5 Jun 2025, Huayra et al., 2023).

Observed ranges:

Channel	$\sigma_\mathrm{eff}$ [mb]	Reference
Four jets (LHC, $\sqrt{s}=7$–13 TeV)	$13$–$16$	(Sadeh, 2013, Collaboration, 2021)
$W$ + 2 jets (LHC, 7 TeV)	$20.7 \pm 6.6$	(Collaboration, 2013)
Same-sign $WW$ (LHC, 8–13 TeV)	$10.6 \pm 1.8$ – $12.2$	(Collaboration, 13 May 2025, Collaboration, 2017)
$J/\psi\,J/\psi$ (LHCb, 13 TeV)	$7$–$8$	(Collaboration, 2021)

Apparent process and kinematic dependence of $\sigma_\mathrm{eff}$ is now well established.

Theoretical explanations for this variation invoke $x$ and scale ($\mu$) dependent transverse profiles in DPDs, as well as parton flavor composition. Global fits (Lovato et al., 5 Jun 2025) confirm that $\sigma_\mathrm{eff}$ is minimized in regions dominated by compact, low-$x$ gluons (forward heavy quarkonia, $\sim2$ mb), and maximized for large-$x$, valence-dominated regions (central high-$p_T$ jets, $15$–$20$ mb), with mild $\mu$ evolution.

4. Theoretical Interpretations and Model Implementations

The fundamental interpretation of $\sigma_\mathrm{eff}$ is as an inverse measure of the partonic overlap in the transverse plane,

$$\sigma_\mathrm{eff}^{-1} = \int d^2\mathbf{b}\,[T(\mathbf{b})]^2,$$

with $T(\mathbf{b})$ the normalized transverse parton density (Rinaldi et al., 2018, d'Enterria et al., 2017). Variants allow for parton flavor (valence vs. sea) and $x$-dependent widths (Huayra et al., 2023). In impact-parameter space, $\sigma_\mathrm{eff}$ relates to the mean pairwise distance $\langle b^2 \rangle$, with rigorous bounds

$$\frac{\sigma_\mathrm{eff}}{3\pi} \leq \langle b^2 \rangle \leq \frac{\sigma_\mathrm{eff}}{\pi}$$

(Rinaldi et al., 2018).

QCD-inspired or phenomenological models extend this picture:

• Light-Front and Holographic QCD: The Light-Front constituent quark and AdS/QCD Soft-Wall models predict explicit $x$-dependence, with $$\sigma_\mathrm{eff}(x_1,x_2) \propto \ln(1/x_1) + \ln(1/x_2)$$, and highlight dynamical two-parton correlations (Rinaldi et al., 2016, Rinaldi et al., 2015, Traini et al., 2016).
• Gaussian and Soft-Gluon Models: Gaussian transverse profiles fit to global DPS data provide compact analytic control, modeling both $x$ and $\mu$ evolution of the effective area (Lovato et al., 5 Jun 2025).
• Process/Flavor Dependence: Studies distinguishing sea–sea versus sea–valence pairings find that sea–sea partons are more tightly localized; process-dependent $\sigma_\mathrm{eff}$ values can be traced to the varying weights of these contributions (Huayra et al., 2023).
• Reggeon Field Theory and Multiparton Correlations: RFT (QGSJET-II) and related approaches model interplay between independent partons, perturbative, and nonperturbative (Pomeron-enhanced) correlations, resulting in a weak $\sqrt{s}$ and $p_T$ dependence and $\sigma_\mathrm{eff}$ consistent with data (Ostapchenko et al., 2015).

5. Extensions to Nuclei and Non-pp Collisions

The extension of $\sigma_\mathrm{eff}$ to proton–nucleus and nucleus–nucleus collisions is well defined. The effective area decreases sharply with increasing nuclear mass due to the enhanced probability of scattering off multiple nucleons:

• For pA: $$\sigma_\mathrm{eff}^{pA} \simeq [A/\sigma_\mathrm{eff}^{pp} + F_{pA}]^{-1}$$, with $F_{pA}\sim30$ mb$^{-1}$ for Pb. For $A=208$ and $\sigma_\mathrm{eff}^{pp}=15$ mb this gives $\sigma_\mathrm{eff}^{pA}\sim22\ \mu$b.
• For AA: $\sigma_\mathrm{eff}^{AA} \sim 1.5$ nb for Pb–Pb (d'Enterria et al., 2014, d'Enterria et al., 2017).

These scaling relations, computed using Glauber models and the geometry of nuclear thickness functions, enable systematic DPS and higher-multiplicity scatter calculations for heavy ion physics.

6. Photon–Proton and Electroweak Variants

In quasi-real photon–proton scattering, the effective cross section generalizes to $\sigma_\mathrm{eff}^{\gamma p}(Q^2)$, with $Q^2$ the photon virtuality. This parameter directly images the convolution of the photon's and proton's transverse profiles. At low $Q^2$, $\sigma_\mathrm{eff}^{\gamma p}$ is large ($\sim30$–$100$ mb), decreasing with increasing $Q^2$ towards typical pp values, and can thereby probe the spatial distribution of partons in the proton in a controlled way (Matteo, 2022, Rinaldi et al., 2021).

7. Phenomenological Implications and Open Issues

• The measured range $8$–$25$ mb for $\sigma_\mathrm{eff}$ across energies and processes implies an average root-mean-square transverse parton separation of $0.2$–$0.9$ fm, smaller than the naïve proton radius and suggestive of transverse correlations or spatial “hotspot” structure (Rinaldi et al., 2018).
• A universal, process-independent $\sigma_\mathrm{eff}$ is no longer tenable; accounting for explicit $x$, $\mu$, and flavor dependences is necessary for quantitative modeling and MC event generator tuning (Lovato et al., 5 Jun 2025, Huayra et al., 2023).
• High-precision studies at the LHC and prospective facilities (EIC, FCC) will further test $x$ and process dependence, and permit mapping of the two-parton transverse correlations within the proton, providing critical tomographic insight beyond standard single-parton imaging.

The effective DPS cross section thus serves as a unique probe of three-dimensional nucleon structure, encompassing both its intrinsic geometry and the emergent QCD correlations at multi-parton level.