Home » Blogs » tnonaka's blog

Centrality definitions (2018~)

Updated on Wed, 2020-01-08 23:25. Originally created by tnonaka on 2019-05-14 14:54.
=====================================================================
 This page describes the information about the centrality definition starting from Run17 data.
 This is based on the discussions here.
 Information about the centrality definitions in BES-I data can be found here.

=====================================================================
 

Run17, 54.4 GeV RefMult
studied by Shaowei Lan, CCNU

Discussions
  • Pile-up rejection by TOF mult. vs RefMult correlation as well as |VpdVz-TpcVz|<3 cm

  • Trigger : 580001, 580021 (580001 is not analyzed due to small statistics)
  • No Luminosity dependence for <RefMult>
  • Different Vz acceptance between two triggers, acceptance correction is not needed for 580021 (|Vz|<40cm)
  • Special treatment is needed for 580001

Max refmult extracted from the error function fitting to the tail of RefMult distribution as a function of Vz, which is fitted by 6th-order polynomial. Red squares are corrected one. Correction is not necessary for trigger 580021.
  • Different shape of RefMult distribution at the edge of Vz
  • See the explanation from Jamie here
    • Online vertex cut was |Vz|<30 cm for 580001, while it was changed to |Vz|<50 cm with requiring TofMult>0.
    • Details can be found in Jamie's blog

  • Centrality resolution is quite different for Vz<-27 and Vz>25 cm. So basically we suggest not to use those Vz region in physics analysis, or simply remove trigger 580001 (much smaller statistics than 580021).
  • MC glauber fits have been performed for those Vz window separately to define the Vz dependent centrality definitions, see slide 13. Previous StRefMultCorr package didn't allow the multiple centrality definition in the same run number, so the codes are modified accordingly. 


Package updates
  • All the correction parameters and centrality definitions in askii files (Centrality_def_(g)refmult(X).txt) are moved to the header file Param.h
  • Bad runs (bad_runs_refmult_year2010(2011)) are merged to BadRun.h
  • In order to use the header file instead of askii file with keeping the current shape of the package as much as possible, some functions are added/modified/replaced in StRefMultCorr class
    • read() -> readHeaderFile()
    • readBadRuns() -> readBadRunsFromHeaderFile()
    • Added StringSplit()
    • Added getRefX()
    • Added getNumberOfDatasets()
  • 54.4GeV RefMult correction updated
  • In order to implement the Vz dependent centrality definitions, several functions are added in StRefMultCorr class
    • getVzWindowForVzDepCentDef()
    • getCentralityBin9VzDep()
    • getCentralityBin16VzDep()
  • Note that those function updates/modifications are blind for users, procedures to use the package don't change. Please leave your macro as it is, just update the package to the latest version.



 
»