Proton Structure

The process of heavy quark production is directly sensitive to the distributions of heavy quarks (massless approach, fig. 2) and/or gluons (massive approach, fig. 3) in the proton. In global fits to inclusive data, the gluon distribution $g(x,Q^2)$ is extracted by analysis of the scaling violations of the proton structure function assuming a certain functional form for $xg(x,Q^2)$. In contrast, heavy quark processes can be used to determine the gluon distribution directly, i.e.by reconstruction of the kinematics of the interacting partons from the measurement of the hadronic final state. Such direct measurements are complementary to the indirect analyses and - although still limited in statistics - they are in principle more sensitive to local variations. In fig. 4 a comparison is shown of the gluon distribution as extracted from global fits and from two sets of charmed $D^{*\pm}$-meson data collected at H1 in DIS and in photoproduction [4]. The gluon density is extracted from the $D^*$ cross section using an unfolding procedure in which effects from gluon radiation and fragmentation are removed. The relative contribution from $D^*$ production via quarks from the proton is subtracted. For the different bins of the measurement the gluon densities are obtained at different factorization scales as given by the phase space of the particular bin and evolved to a scale $\mu^2=25$ GeV$^2$.

Figure 4: Gluon distributions $xg(x,Q^2)$ as extracted from $D^*$ data [4]. The systematic error is a quadratic sum of all contributions and is dominated by the theoretical uncertainty on the charm quark mass (DIS sample) and the renormalization and factorization scale (photoproduction).