MadGraph5_aMC@NLO

What is the connection between interference in integration channels and diagrams in MadGraph 5?

Asked by Tathagata Ghosh

Hi Madgarph experts,

I was going through inner workings of madgraph5 and came accross the nice expalanation in the link "https://answers.launchpad.net/mg5amcnlo/+question/187387". I now understand the workings of MG5 much better but still one thing is not clear to me. I'm wondering how can I found the cross-section of the cross-channel interference diagrams?

 For example if I generate "p p > e+ e-" then G1, which is s-channel gamma diagram gives me 207.1 fb and G2, which is s-channel Z diagram gives me 639.4 fb. However there should be an interference term also. How MG5 take into account that contribution and how can I read that number from the files generated?

Sincerely,
Tathagata Ghosh

Question information

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

Hi,

From those number you can not infer the value of the interference term. Both G1 and G2 include the interference term in the computation since G1 correspond to integrate the following function:

|M_1|^2 / (|M_1|^2+|M_2^2|) * |M_1+M_2|^2

While G2 correspond to integrate
|M_2|^2 / (|M_1|^2+|M_2^2|) * |M_1+M_2|^2

So as you should have G1+ G2 correspond to integrate
|M_1+M_2|^2
 which includes the interference as it should.

If you are interested in interference term, it is now possible to compute the interference term automatically between different order.
The syntax is the following
generate p p > j j QED^2==2

If you want to have the pure QCD contribution + the interference with the QED term then you can do:
generate p p > j j QCD^2<=2

So with this syntax, we can not do the interference between the Z/a propagator out of the box. On the otherhand nothing prevent you to change the Z coupling from QED to something else and use the mentioned syntax.

Cheers,

Olivier

On 19 Mar 2015, at 18:16, Tathagata Ghosh <email address hidden> wrote:

> New question #263848 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/263848
>
> Hi Madgarph experts,
>
> I was going through inner workings of madgraph5 and came accross the nice expalanation in the link "https://answers.launchpad.net/mg5amcnlo/+question/187387". I now understand the workings of MG5 much better but still one thing is not clear to me. I'm wondering how can I found the cross-section of the cross-channel interference diagrams?
>
> For example if I generate "p p > e+ e-" then G1, which is s-channel gamma diagram gives me 207.1 fb and G2, which is s-channel Z diagram gives me 639.4 fb. However there should be an interference term also. How MG5 take into account that contribution and how can I read that number from the files generated?
>
> Sincerely,
> Tathagata Ghosh
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Tathagata Ghosh (ghoshtatha) said : 		#2

That answers my question.