Shadow generalized parton distributions: a practical approach to the deconvolution problem of DVCS

TL;DR

This paper presents a practical method for deconvolving generalized parton distributions from DVCS data, highlighting the limitations of DVCS alone for model-independent GPD extraction and emphasizing the need for multi-channel analysis.

Contribution

It introduces a next-to-leading order analysis approach and demonstrates that shadow GPDs can have minimal impact on observables, challenging the feasibility of model-independent GPD extraction from DVCS.

Findings

Shadow GPDs can be made arbitrarily small in DVCS observables.

DVCS alone cannot provide a model-independent GPD extraction.

Multi-channel analysis is necessary for reliable GPD determination.

Abstract

Deeply virtual Compton scattering (DVCS) attracts a lot of interest due to its sensitivity to generalized parton distributions (GPDs) which provide a rich access to the partonic structure of hadrons. However, the practical extraction of GPDs for this channel requires a deconvolution procedure, whose feasibility has been disputed. We provide a practical approach to this problem based on a next-to-leading order analysis and a careful study of evolution effects, by exhibiting shadow GPDs with arbitrarily small imprints on DVCS observables at current and future experimental facilities. This shows that DVCS alone will not allow for a model independent extraction of GPDs and a multi-channel analysis is required for this purpose.