P oS(DIS 2010)053 The Effect of Variable Flavour Number Scheme Variations on

I consider variations in the denition of a general-mass variable avour number scheme (GM-VFNS) for heavy avour structure functions, both at next-to-leading order (NLO) and at next-to-next-to leading order (NNLO).I also dene a new optimal scheme choice improving the smoothness of the transition from one avour number to the next. At both NLO and NNLO I investigate the variation of the structure function for a xed set of parton distribution functions (PDFs) and also the change in the distributions when a new MSTW-type global t to data is performed for each GM-VFNS. At NLO the parton distributions, and predictions using them at hadron colliders, can vary by 2 % from the mean value. Use of the the zero-mass variable avour number scheme, which is simpler but only an approximation, leads to results a further couple of percent or more outside this range. At NNLO there is far more stability with varying GM-VFNS denition. Typical changes in PDFs and predictions are less than 1%, with most variation at very small x values. This demonstrates that mass-scheme variation is an additional and signicant source of uncertainty when considering parton distributions, but like other theoretical uncertainties, it diminishes quickly as higher orders are included