- Slides: 7
Calibration hits in Jets A proposal for a package analyzing calibration hits inside reconstructed jets Pierre–Antoine Delsart Mohsen Kharzad & Rachid Mazini
Package Overview Results from the code being written forward jets analysis ●Quite generic thanks to back navigation ⇒ can be useful to others ●Consists of one simple analysis algorithm ● It uses fully simulated events with calibration hits, produces an ntuple & associates to each jet : ➢Sum ➢An of energy of the hits inside the jets array corresponding to some calorimeter bins holding the jet energy deposition in each bin This for each 'type' of energy (G 4 Total, G 4 visible, etc. . . ) Calo is divided in : ●geometric region (Barrel, Tile, HEC, . . . ) ●Energy density bins (1 or 16) as in current hadronic calibration
Package Details - Inputs It requires ESD files containing Calo. Cell. Containers filled with Calibration hits. Simulation with calibration hits Digitization + keep calib hits Add this line in job. Option : Stream 1. Item. List+=["Calo. Calibration. Hit. Container#*"] Reconstruction + Calib. Hit. To. Calo. Cell algo - Add job. Option : “Calib. Hit_joboption. py” - May add 'topo jets' ESD file
Package Details - algo In each event, for each jet : Non. Em Energy Jets Visible Energy Total Energy G 4 Calo. Cell. Container back navigation reco cells Identifier. Hash G 4 cells E+=Ecell ●Find and fill calo bin according to ● Calo. Sample and E density Other convenient calculation. . . ●
Algorithm Output – Analysis script The algo produces an ntuple with jet related info : - Jet E, p. T, eta, phi (for jets before and after calibration) - same for (eta, phi)-matched truth jet - G 4 Total, visible, Non. Em energy of the jet - array of calo bins in the form 'Zone. Energy[njet][nbin]' for reco and all G 4 energies - Radial energy profile of the jet The granularity of calo bins is quite fine (505 bins) : allows flexible analysis but information extraction is not immediate A ROOT script is provided : allows to easily convert the initial bin division into - Hadronic weight division - Calo Sample division - custom division. . .
Analysis examples (di-jet sample, E~1 Tev, eta>3) Ratio E/Etruth in E density bins Total jet energy/ True jet energy distribution Reco/True G 4 / True G 4 total G 4 Visible G 4 non EM Reco Jets in HEC-FCAL crack Hadronic weight binning Ratio G 4/Reco uncalibrated Calibration weights
Work in progress – open questions ● ● Dead Material hits information is being implemented in the algo : - by building a (eta, phi) distribution of DM energy - associating each jet to a (eta, phi) DM subregion Currently having problems with cal hits in 10. 0. 4 (wrong Ids). Being fixed. . . If interest for this analysis algorithm, it is ready to be released as a pakage - New package (where ? ) or add it to an existing package ? - Easy to re-implement it as an Athena tool or algo if prefered - Easy to extend it to any type of calorimeter object (back navigation) A test version is available here : /afs/cern. ch/user/d/delsart/publi/Jet. Analysis. tar. gz Please email me for details (delsart at lps umontreal ca)