MadGraph5_aMC@NLO

Meaningful cross-sections of individual diagrams?

Asked by Emanuel Gouveia

Dear experts,

I generated the LO process
p p > t t~ x0
with the Higgs Characterisation UFO model using MG5 version 2.6.5

In the results page, I find cross-section contributions of the integration channels (Gs) for the gg initiated subprocess.
I'm not sure to which extent those could be interpreted as contributions of individual diagrams to the cross-section.

I read the answers in https://answers.launchpad.net/mg5amcnlo/+faq/2944 and https://answers.launchpad.net/mg5amcnlo/+question/321938, from which I realise that this interpretation is not in general possible.

What made me wonder about it being possible in this particular case is the content of config_subproc_map.inc:
  DATA (CONFSUB(I,1),I=1,1)/1/
  DATA (CONFSUB(I,2),I=1,1)/2/
  DATA (CONFSUB(I,3),I=1,1)/3/
  DATA (CONFSUB(I,4),I=1,1)/4/
  DATA (CONFSUB(I,5),I=1,1)/5/
  DATA (CONFSUB(I,6),I=1,1)/6/
  DATA (CONFSUB(I,7),I=1,1)/7/
  DATA (CONFSUB(I,8),I=1,1)/8/

So, there seems to be a one-to-one correspondence between Gs and diagrams, right?
Does this make the cross-section contributions from individual Gs in anyway interpretable as contributions from diagrams?

Best,
Emanuel

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
Last query:
Last reply:
Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#1

Hi,

In some sense this does not help that much, you still have the interference term which is not clearly assigned to any of the channel of integration. This should be consider at best as an indication of the importance of the various diagram not as the contribution of a given diagram (in top of that this is not guarantee to be gauge invariant)

In practice, those number are mainly useful for debugging/ checking that everything is fine.

Cheers,

Olivier

Revision history for this message
Emanuel Gouveia (eddygouveia91) said : 		#2

Thanks Olivier Mattelaer, that solved my question.