Jets faking photons

[Matias Vazquez]

Dear Delphes authors,

I am trying to simulate a background for a b b~ gamma gamma final state, for an study on double Higgs production. One of the important sources in this background is the b b~ j gamma, where the jet is mistagged as a photon. Up to now, I was generating events with SHERPA and then running the detector simulation using Delphes with the default ATLAS card, where I have only changed the parameters for the b-tagging efficiencues (delphes 3.3.2. find attached the card). I saw in this ticket,
https://cp3.irmp.ucl.ac.be/projects/delphes/ticket/725

that one can use 'JetFakeParticle' to fake photons originating from jets. My questions are:

1) Is it enough if I simply add all the JetFakeParticle lines related to photons as in the test_v3.tcl in ticket #725?

2) Are there any numbers for these mistagging efficiencies from the experimental collaborations? Which are realistic numbers for the jets-faking-photons rate?

Many thanks in advance for your help,
Matias.

[Michele Selvaggi]

Hi,

1) yes
2) These highly depend on isolation/ identification criteria, so you'll have to check with your experimental colleagues. Also keep in mind that the jet fragmentation naturally produces lots of photons through pi0 decays, and some of those will be isolated. So i think you first want to check how many jets produce isolated photons without applying a fake rate and then eventually apply one. Also, maybe a good idea would be to compare different hadronization models (eg sherpa vs. pythia8)

Michele

[Matias Vazquez]

Hi Michele,

Thanks for your detailed and fast reply. I will do that.

Cheers,
Matias.

Replying to mselvaggi:

Hi,

1) yes
2) These highly depend on isolation/ identification criteria, so you'll have to check with your experimental colleagues. Also keep in mind that the jet fragmentation naturally produces lots of photons through pi0 decays, and some of those will be isolated. So i think you first want to check how many jets produce isolated photons without applying a fake rate and then eventually apply one. Also, maybe a good idea would be to compare different hadronization models (eg sherpa vs. pythia8)

Michele