Initial condition for the Balitsky-Kovchegov equation at next-to-leading order

Carlisle Casuga, Henri Hänninen, Heikki Mäntysaari · 2025

Read Initial condition for the Balitsky-Kovchegov equation at next-to-leading order by Carlisle Casuga, Henri Hänninen, Heikki Mäntysaari (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-00487.txt.

Introduction

Carlisle Casuga, Henri Hänninen and Heikki Mäntysaari fix the starting state of a key equation for high-energy nuclear physics, the Balitsky-Kovchegov equation, at next-to-leading order. The equation tracks how gluon density grows as particles are probed harder, inside the colour glass condensate frame. Its starting shape was a free input, and its error bars matter for every later result.

Fitting HERA collider data on total cross sections and charm output pinned the start down tightly. Bayesian inference gave full probability ranges for the starting shape, not just one best fit. Those ranges now let every next-to-leading calculation carry honest uncertainty from the start.

Catalog record

CREATORCarlisle Casuga, Henri Hänninen, Heikki Mäntysaari
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE46 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-00487

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Initial condition for the Balitsky-Kovchegov equation at next-to-leading order Carlisle Casuga, Henri Hänninen, Heikki Mäntysaari Abstract We determine the initial condition of the Balitsky-Kovchegov evolution equation at next-to-leading order (NLO) accuracy using HERA deep inelastic scattering data. Posterior distributions characterizing the initial condition are extracted using Bayesian inference. The total cross section and charm quark production data from HERA are found to provide stringent constraints on the posterior. These distributions quantify the uncertainty in the initial condition and serve as necessary input for propagating uncertainties to all NLO calculations within the Color Glass Condensate framework. I Introduction The Color Glass Condensate (CGC) Iancu:2003xm effective field theory provides a natural framework for describing the gluon-dense, high-energy structure of protons and nuclei. As the collision energy increases or the Bjorken- x decreases at fixed probe virtuality Q^{2} , one unlocks the saturation regime with emergent non-linear QCD effects. This regime is characterized by the saturation scale, Q_{s}^{2} , which scales as Q_{s}^{2}\sim A^{1/3} , where A is the nuclear mass number. In this context, the upcoming Electron Ion collider (EIC) AbdulKhalek:2021gbh (or other potential future nuclear-DIS facilities LHeC:2020van ; Anderle:2021wcy ) will feature heavy nuclear targets and provide access to the saturated state of gluonic matter. In the leading order (LO) dipole picture of deep inelastic scattering (DIS), the incoming virtual photon splits into a quark-antiquark pair which interacts with the target nucleus Mueller:1994jq . The necessary ingredients in calculations of DIS structure functions in the CGC framework are the DIS impact factor and the small- x Balitsky-Kovchegov (BK) Kovchegov:1999yj ; Balitsky:1995ub evolution equation at the desired order in perturbation theory. First principle calculations of impact factors at NLO in \alpha_{\mathrm{s}} have been performed in the light cone perturbation theory for light Balitsky:2010ze ; Balitsky:2012bs ; Hanninen:2017ddy ; Beuf:2017bpd ; Beuf:2016wdz ; Beuf:2011xd and massive Beuf:2022ndu ; Beuf:2021qqa ; Beuf:2021srj quarks. These impact factors, which describe the photon splitting into partonic states, are coupled to the high-energy BK evolution equation that resums contributions enhanced by large logarithms of energy \sim\alpha_{\mathrm{s}}\ln(1/x) (or \alpha_{\mathrm{s}}^{2}\ln(1/x) at NLO). The full NLO BK evolution equation is derived in Ref. Balitsky:2007feb with numerical solutions implemented in Refs. Lappi:2015fma ; Lappi:2016fmu . Prescriptions that resum dominant single and double logarithmic contributions, that provide an accurate approximation to the NLO BK equation Lappi:2016fmu , have been obtained in Refs. Ducloue:2019ezk ; Iancu:2015vea ; Iancu:2015joa ; Beuf:2014uia . The BK equation describes the energy dependence of the dipole-target scattering amplitude. This perturbative evolution equation takes as a non-perturbative input the dipole-target scattering amplitude at an initial energy scale, and predicts its energy evolution towards higher energies. Models like the McLerran-Venugopalan (MV) McLerran:1993ni model or the Golec-Biernat-Wüsthoff (GBW) model GolecBiernat:1998js provide reasonable parametrizations for this initial condition. Substantial efforts have been made in extracting the initial condition from available data at LO Albacete:2009fh ; Albacete:2010sy ; Lappi:2013zma , and at NLO with light Beuf:2020dxl and heavy Hanninen:2022gje quarks. In particular, Ref. Hanninen:2022gje shows the feasibility of a global fit at NLO accuracy by demonstrating that it is possible to simultaneously describe both the total cross section H1:2015ubc and heavy quark production H1:2018flt data from HERA at small x . These previous NLO dipole fits, however, do not include the heavy quark contribution in the actual fit procedure, and neglect correlated experimental uncertainties. Furthermore, they also lack a systematic quantification of the uncertainty of the determined non-perturbative BK initial condition, which has been obtained in a leading order analysis in Ref. Casuga:2023dcf . An accurate extraction of this non-perturbative input, among with well-defined uncertainty estimates, is essential in order to promote the precision in theoretical calculations to the level that is comparable to current and future high-precision measurement sensitive to saturation effects at the EIC or at the LHC Morreale:2021pnn . The main objective of this work is to extend our previous analysis Casuga:2023dcf to infer the initial condition for the BK equation in NLO accuracy. For the first time, we perform a global analysis including both the HERA total cross section and the charm production data as constraints. To extract the model parameters in the initial condition, we use Bayesian inference, which not only identifies the best-fit values but also probes its neighboring area in the parameter space. This approach determines posterior distributions that provide a streamlined method to propagate the initial condition uncertainty to calculations of other observables within the CGC framework at NLO. The paper is structured as follows. Sec. II provides a brief summary of the DIS cross section calculation at the next-to-leading order accuracy in the CGC approach. This involves a two-part discussion of the NLO impact factors in Sec. II.1 and the approximated NLO evolution equation in Sec. II.2 . Details of the Bayesian inference setup used to find the probability distribution describing the initial condition to the BK evolution equation are presented in Sec. III . The obtained posterior distributions, best-fit values, and other results are then shown in Sec. IV . Conclusions and main findings are summarized in Sec. V . II Deep inelastic scattering at NLO in dipole picture II.1 NLO cross section The leading order dipole picture of DIS consists of a photon fluctuating to a dipole and interacting instantaneously with the target hadron by exchanging low- x gluons. The separation between the lifetime of the dipole and the timescale of the interaction enables one to factorize the leading order \gamma^{*}A cross section, as a function of Bjorken- x and photon virtuality, Q^{2} , as Kovchegov:2012mbw \sigma^{\gamma^{*}p}_{T,L}(x,Q^{2})=2\sum_{f}\int\differential[2]{\bt}\differential[2]{\rt}\differential{z}|\psi^{\gamma^{*}\to q\bar{q}}_{T,L}(\mathbf{r},Q^{2},z)|^{2}\\ \times N_{\textrm{LO}}(\mathbf{r},\mathbf{b},x). (1) The light-cone wave function, \psi^{\gamma^{*}\rightarrow q\bar{q}}_{T,L} , describes the splitting of the virtual photon to the quark-antiquark pair with a transverse separation \mathbf{r} , where the quark carries a fraction, z , of the large photon plus momentum. Here T and L refer to the transverse and longitudinal photon polarization, respectively. The dipole interacts with the target color field eikonally, so that \mathbf{r} does not change during the interaction. In this work we further neglect the impact parameter dependence from the dipole-target scattering amplitude N_{\textrm{LO}} and, following e.g. Refs. Albacete:2010sy ; Lappi:2013zma , replace \int\differential[2]{\bt}\to\sigma_{0}/2 where \sigma_{0}/2 is the proton transverse area and \mathbf{b} is the impact parameter in the dipole-target scattering. The total DIS cross section data is typically reported as a reduced cross section defined as \sigma_{r}(y,x,Q^{2})=F_{2}\left(x,Q^{2}\right)-\frac{y^{2}}{1+(1-y)^{2}}F_{L}\left(x,Q^{2}\right). (2) Here the inelasticity is defined as y=Q^{2}/(sx) , where \sqrt{s} is the center-of-mass energy in the electron-proton scattering. The structure functions F_{2} and F_{L} are related to the virtual photon-proton cross section as \displaystyle F_{2} \displaystyle=\frac{Q^{2}}{4\pi^{2}\alpha_{\mathrm{em}}}(\sigma^{\gamma^{*}A}_{T}+\sigma^{\gamma^{*}A}_{L}), (3) \displaystyle F_{L} \displaystyle=\frac{Q^{2}}{4\pi^{2}\alpha_{\mathrm{em}}}\sigma^{\gamma^{*}A}_{L}. (4) The NLO corrections to the \sigma^{\gamma^{*}A}_{L,T} cross section include the tree-level contribution describing the photon fluctuating to a q\bar{q}g system, where the additional gluon also interacts with the target. Furthermore, NLO corrections also include virtual one-loop corrections to the light-cone wave function describing the \gamma^{*}\rightarrow q\bar{q} splitting. The q\bar{q}g and the one-loop q\bar{q} contributions in the dipole factorization have been first calculated in the massless limit in Refs. Hanninen:2017ddy ; Beuf:2016wdz ; Beuf:2017bpd , and later results with a finite quark mass enabling also calculations of heavy quark production were published in Refs. Beuf:2022ndu ; Beuf:2021srj ; Beuf:2021qqa . At NLO accuracy, a rapidity divergence appears when the gluon longitudinal momentum

Read the complete text