WHIZARD

Photon matching and LHEF output

Asked by Suchita

Dear WHIZARD authors,

I am trying to generate a LHEF file using WHIZARD 3.1.4 where I would like to eventually implement photon ME and ISR photon matching procedure. In this context I have two questions which I detail below.

First, I would like to know if the resulting LHEF output should respect the 'cuts' defined in the .sin file?

As a concrete example, I am using a simplified version of example from here: https://indico.cern.ch/event/932973/contributions/4056012/attachments/2140169/3606831/FCC-JK.pdf (Just to make sure that I am not using wrong commands).

sqrts = 3000 GeV
n_events = 1000
checkpoint = 1000
mH = 12 TeV
alphas = 0 # to remove overlap with higher multiplicities
show (model)
alias nu = n1:n2:n3
alias NU = N1:N2:N3
process eevv = e1, E1 => (nu, NU)
process eevv1a = e1, E1 => (nu, NU, A)
process eevv2a = e1, E1 => (nu, NU, A, A)
cuts = let subevt @meA = select if Index > 2 [A]
       in all 2*sqrt(1500*E)*sin(Theta/2.) > 10 GeV [@meA]
       and all 2*sqrt(1500*E)*cos(Theta/2.) > 10 GeV [@meA]
compile
beams = e1, E1
integrate (eevv,eevv1a,eevv2a) {relative_error_goal=0.01 iterations = 20:20000:"gw", 5:50000:""}
sample_format = lhef
simulate (eevv, eevv1a, eevv2a)
compile_analysis

Would the resulting LHEF file contain events where all photons have q+ and q- > 10 GeV? (q- = 2*sqrt(1500*E)*sin(Theta/2.) and q+ = 2*sqrt(1500*E)*cos(Theta/2.) )

Second: In the above example, I deliberately do not generate any ISR photons. If I add ISR photon to this .sin file, would the resulting LHE file contain correctly matched ME and ISR photons? Would I have to do something in addition? I guess, I am not entirely clear about what exactly gets written in the LHE file.

My .sin file including ISR photons would look like following:

sqrts = 3000 GeV
n_events = 1000
checkpoint = 1000
mH = 12 TeV
alphas = 0 # to remove overlap with higher multiplicities
show (model)
alias nu = n1:n2:n3
alias NU = N1:N2:N3
process eevv = e1, E1 => (nu, NU)
process eevv1a = e1, E1 => (nu, NU, A)
process eevv2a = e1, E1 => (nu, NU, A, A)
cuts = let subevt @meA = select if Index > 2 [A]
       in all 2*sqrt(1500*E)*sin(Theta/2.) > 10 GeV [@meA]
       and all 2*sqrt(1500*E)*cos(Theta/2.) > 10 GeV [@meA]
compile
beams = e1, E1 => isr
?isr_handler = true
$isr_handler_mode = "recoil"
?keep_beams = true
?keep_remnants = true
isr_mass = me
integrate (eevv,eevv1a,eevv2a) {relative_error_goal=0.01 iterations = 20:20000:"gw", 5:50000:""}
selection = all 2*sqrt(1500.*E)*sin(Theta/2.) < 10 GeV [extract index 1 [A]] and
   all 2*sqrt(1500.*E)*cos(Theta/2.) < 10 GeV [extract index 2 [A]]
sample_format = lhef
simulate (eevv, eevv1a, eevv2a)
compile_analysis

I thank you for your help on this and I apologize in advance for these rather trivial questions.

best,
Suchita.

Question information

Language:
English Edit question
Status:
Solved
For:
WHIZARD Edit question
Assignee:
Krzysztof Mekala Edit question
Solved by:
Krzysztof Mekala
Solved:
Last query:
Last reply:
Revision history for this message
Krzysztof Mekala (krzysztofmekala) said : 		#1

Dear Suchita,
the procedure described in the slides and the original paper [2004.14486] corresponds to the case when both ISR and ME photons are generated. First of all, I would recommend referring to the paper, as the example presented there is slightly simpler.

The "cuts" statement affects only particles generated at the Matrix Element level. Thus, the inner "select" statement is used to take into account only the photons with indices greater than 2 (as the ISR photons would be numbered "1" and "2" and the ME photons "3", "4", ...). In your first example, you do not generate any ISR photons and only up to 2 ME photons (which are then numbered "1" and "2" instead) which means no photon has an index greater than 2 and the cuts are effectively not applied. In your second example, when ISR is switched on, you generate two ISR photons, one for each beam ("1" and "2", to which the "selection" statement is applied) and some other ME photons ("3", "4", ..., to which the "cuts" statement is applied) and the matching procedure should work fine. In your LHEF files, you should find beam particles, ISR photons, beam particles after the ISR emission and finally, the particles produced at the ME level.
Also, the "compile_analysis" statement at the end is redundant, as you do not use any internal Whizard analysis tools (histograms, plots, etc.).

I hope this answers your question.

Cheers,
Krzysztof

Revision history for this message
Suchita (suchita-kulkarni) said : 		#2

Dear Krzysztof,

Thank you for the quick answer. From this I understand that the LHE file will contain correctly matched photons, I could just use this file for further analysis without any worries. It also means that in general events in the LHE file obey cuts prescribed in 'cuts' and 'selection' of the .sin file.

If this is correct then my question is for now answered and I do not need any further help.

best,
Suchita.

p.s. I agree that the example in original paper is simpler but it did not contain any scripts for verifying results e.g. by making plots. That made me very confused as I could not know how to establish if I have done everything correctly. My current strategy is to try to reproduce the results from the final slide which should be done using the script given on the slide before. I expect if I did everything correctly, my LHE plots, WHIZARD plots and plots on that slide to be the same.

Revision history for this message
Best Krzysztof Mekala (krzysztofmekala) said : 		#3

Dear Suchita,
yes, your file should contain correctly matched photons and you can use it for further analysis assuming the matching scale between the ISR and ME approaches is suitable for your problem. The main idea behind this matching is that we want to use only "hard" ME photons at the analysis level (as their multiplicities, energies, momenta... are physical) but we also want to account for soft, invisible (no photon experimentally measured) emissions. In case some ISR photons can still be measured in the detector you consider, you should adjust the scale to a lower value (for intuition, see e.g. Fig. 3 in [2004.14486]). If your experimental setup is the same as in the original paper, you can use the same set of cuts of course.

Cheers,
Krzysztof

Revision history for this message
Suchita (suchita-kulkarni) said : 		#4

Thank you, that explains everything I needed.

Revision history for this message
Suchita (suchita-kulkarni) said : 		#5

Thanks Krzysztof Mekala, that solved my question.