implementing cuts on photons at analysis and reconstruction level

[J Dutta]

Hi,

I am studying multiplicity distributions of photons in a BSM model. I expect to see a peak at 2 photons. Using the default delphes_card_ATLAS.tcl and writing a code similar to Example2.C with photon pt's > 10 GeV I see the peak is at 1 photon. I increase this pt cut to higher values 50,100 to see no change in the multiplicity distribution. However I change the photon ptmin cut to 50 in the delphes card. Analysing this root file with the same code gives the multiplicity peak at 2.

I am quite confused as to why the cuts at my analysis level are not reproducing the same behaviour as in the reconstruction level. The cuts seem to be working only at the hep->root conversion level. I have tried changing the isolation variables DeltaRMax, SumPTMax, jet pt cuts. However the same behaviour remains.

Any suggestions in this regard will be very helpful for me to understand and sort this issue. I am using v3.3.2 of Delphes.

Regards,

Juhi

[Michele Selvaggi]

Maybe there is a mistake in your code?
Could you attach a small event sample (~1K events) and the macro you are using to analyse the root file?

[J Dutta]

Hi,

I have attached the sample hep file and the macro (with a pt cut 50 for photon) Please have a look.

Regards,

Juhi

[Michele Selvaggi]

What do you call the ptmin cut?
The one in the isolation module?
This is a cut of the particle considered for computing the isolation, not the photon pt min.
If you set it to 50 you are basically counting no particles in the cone, which results in all photons being isolated.

[J Dutta]

Are you referring to this block in the delphes_card?
##################
# Photon isolation
##################

module Isolation PhotonIsolation {

set CandidateInputArray PhotonEfficiency/photons
set IsolationInputArray EFlowMerger/eflow

set OutputArray photons

set DeltaRMax 0.5

set PTMin 0.5

set PTRatioMax 0.12

}

Indeed I had mistook PTMin 0.5 for photon pt min cut. Thank you for the clarification. Where can I set the photon pt min or lepton pt min cut directly in the card?

However I had kept PTMin 0.5 in the first case in the card. In the macro I am placing extra cuts on photon pt as 10,50 to identify photons. In this case when I am increasing the photon pt cut in the macro to 50,100 and not in the card I see a peak at 1 whereas i expect it at 2 which I see at the pythia level but not in delphes level.

Regards,

Juhi

[Michele Selvaggi]

here:

###################
# Photon efficiency
###################

module Efficiency PhotonEfficiency {
  set InputArray Calorimeter/eflowPhotons
  set OutputArray photons

  # set EfficiencyFormula {efficiency formula as a function of eta and pt}

  # efficiency formula for photons
  set EfficiencyFormula {                                      (pt <= 50.0) * (0.00) +
                                           (abs(eta) <= 1.5) * (pt > 10.0)  * (0.95) +
                         (abs(eta) > 1.5 && abs(eta) <= 2.5) * (pt > 10.0)  * (0.85) +
                         (abs(eta) > 2.5)                                   * (0.00)}
}

however I suggest you apply the cut at the analysis stage.

Your photons are very energetic, this is why the pt cut does not affect the multiplicity distribution.

You can see that:

root file.root
root [1] Delphes->Draw("Photon.PT")

[J Dutta]

yes the photons are very energetic which is why i was expecting the multiplicity to peak at 2 always and not at one. The fact that so many photons are failing the isolation cut is very strange to me. Am I making an error at some stage? Is there any way I can correct this or check for consistency in this case?

Regards,

Juhi

[Michele Selvaggi]

You can try to put the efficiency formula above to 100%. Although that will be probably be over optimistic, it will be closer to reality given that high pt photons are very likely to be reconstructed.

This said you should check numbers published by ATLAS. The values in the card are simply indicative.

[J Dutta]

Thank you for the suggestions. I will check them and let you know in case I have any further queries.

Thanks & Regards,

Juhi