Phenomenology of Deep Inelastic Scattering Structure Functions

I review recent progress in analysing deep inelastic scattering structure functions in global analyses. The new ingredients are new data and attempts to incorporate heavy quarks consistently. A new way of including the resummation of large log 1/x terms is discussed. INTRODUCTION Since we last met in Rome last year there has been progress in our understanding of deep inelastic scattering (DIS) structure functions and the implications of new measurements in the context of global analyses and extraction of parton distribution functions (pdf’s). Firstly there have been updates to two major experiments − the muon-nucleon NMC collaboration at CERN and the neutrino-iron CCFR collaboration at Fermilab. The NMC [1] has produced their final numbers for F2 for proton and deuterium targets at 90, 120, 200 and 280 GeV while the CCFR [2] collaboration has implemented new energy calibrations to its analysis. The latter data are consistent with a greater value of αs(MZ) than the previously reported value. Finally there have been the first reported measurements from HERA on the charm structure function F c 2 [3,4] which indicate that a substantial fraction of the total F2 at HERA comes from charm production. On the theoretical side, progress has been made following improved treatments of heavy quark production in DIS. Why is this important? From above we see that the component of F2 is relatively large and so it is obviously important to have a consistent description of this component and to check this with boson-gluon fusion mechanisms. Also if we are trying to understand the small x behaviour of F2 (i.e. BFKL versus GLAP) then the data force us to study small Q i.e. the region where we move through the charm threshold; so this should be understood. Finally 1) Plenary Talk at 5th International Workshop on Deep Inelastic Scattering and QCD (DIS’97), Chicago, April 1997 it is necessary to produce pdf’s for charm and bottom flavours to insert into other processes such as jet production. Large x, high Q is where the excitement resides just now. Can we be precise about the conventional background and estimate the uncertainty in the cross section around Q ∼ 10GeV , x ∼ 0.45? I shall discuss this briefly. For some time it has been realised that in addition to resumming large logs in Q (GLAP), there may be significant contributions arising from resumming potentially large logs in 1/x (BFKL). The difficulty is to devise a theoretically consistent procedure of ‘ marrying’ the two in a practical way that confronts the data and tries to answer the question whether these data favour or disfavour the inclusion of the log 1/x resummed terms. Here there has been definite progress and I shall highlight a recent analysis by Thorne [5]. RECENT GLOBAL ANALYSES The two major providers of pdf’s, CTEQ and MRS, continue to update their global analyses of DIS and other processes. Both are now attempting to include a theoretically improved treatment of charm and bottom and these treatments are discussed in the following section. In this section I will highlight some results emerging from these on-going analyses. Lai and Tung [6] demonstrate the effect 10-4 10-3 10-2 X 0.6 0.8 1 1.2 1.4 1.6 F 2 H1: Q2 = 5GeV2 H1: Q2 = 15GeV2 H1: Q2 = 25GeV2 CTEQ4M CTEQ4M (massive) CTEQ4HQ FIGURE 1. Comparison of H1 data [8] with calculations using CTEQ4M pdf’s in the original massless scheme (solid line) with the new fit CTEQ4HQ using the ACOT procedure (dotted line). of including the variable flavour scheme (VFNS) developed by Aivazis et al. [7] (ACOT) based on their ealier work [9] into the CTEQ analysis. This is to be contrasted with the previous treatment in CTEQ where charm evolved as a massless quark. The effect is illustrated in fig. 1. Only a small change is observed as one changes from CTEQ4M (the massless evolution) to the improved VFNS procedure which is more noticeable at small x. A comparison of the resulting pdf’s at Q = 25 GeV is shown in fig. 2 Moreover, 10-4 10-3 10-2 10-1 100 X 0 0.5 1 1.5 x * f ( x, Q = 5G eV )