Update – Jet Measurements to be used in PDF fits

ratio of measured cross section to NLO pQCD prediction, in the high-rapidity region

Yesterday I wrote about the CDF di-jet mass spectrum and suggested that the measurement could be used to improve PDF’s. Ken Hatakeyama (CDF/Rockefeller) pointed out that a recent detailed CDF measurement of jet production will be used to improve PDF’s. The relevant paper is Measurement of the Inclusive Jet Cross Section (Phys. Rev. D78, 052006, 2008). CDF report measurements of differential cross sections as a function of jet E_T for five ranges of rapidity.

The paper provides very clear explanations of the basics of jet measurements: Section III discusses jet reconstruction algorithms, pointing out the differences between cone-based and kT-type algorithms. Section V covers jet energy corrections, which, as I pointed out in the earlier post, are crucial for this kind of analysis. If you want a readable account of those corrections, see this section.

The measured cross sections are slightly lower than predicted. A quantitative test of the standard model predictions (NLO pQCD) was performed based on a chi-squared which incorporated systematic uncertainties and their correlations (Section VIII). The probability of the chi-squared is 4% which does not rule out the standard model. It does suggest that these data might help adjust the PDF’s, a point made in Section VIII. The forward (high rapidity) regions are particularly useful in this regard, especially in connection to the gluon distribution function at high x.

The CDF public web page for this measurement is here.

D-zero published similar measurements somewhat earlier: Measurement of the inclusive jet cross-section (Phys.Rev.Lett. 101, 062001, 2008) and they observed similar trends at high rapidity:

ratio of the D0 measured cross section to NLO pQCD prediction, in the high-rapidity region

The D-zero paper, however, does not describe any discrepancies, instead stating that in all rapidity (y) regions, the predictions agree well with the data. D-zero and CDF use different PDF’s for their analyses (D-zero uses CTEQ6.5 while CDF uses CTEQ6.1M).

The D-zero web page for this analysis is here.