On the Model Independent Extraction of Compton Form Factors from Deeply Virtual Compton Scattering

The deeply virtual Compton scattering process e+p → e+p+γ is a flagship experiment for studying nucleon structure through its dependence on generalized parton distributions. Meanwhile, its cross sections depend directly on an intermediate observable quantity known as Compton form factors (CFFs). I summarize the exercise of locally extracting the real and imaginary parts of the four twist-2 CFFs which arise in this important process. It is found that there are a sufficient number of DVCS observables and degrees of freedom to extract all the twist-2 CFFs model-independently, exploiting the azimuthal dependence of the absolute cross sections across all possible beam and target polarizations at a common kinematical point in {Q,t,xB,y}. To achieve this, re-weighted DVCS cross sections have been used which have also been reduced to their dominant terms, found after averaging them over typical JLab fixed-target kinematics. I demonstrate with both real and pseudo-data how one can extract the twist-2 CFFs using harmonic fitting to both cross sections and asymmetries. This analysis is extended to the twist-3 CFFs as well and the phenomenology of future experimental measurements is discussed.