WW scattering event generation

Hi Niklas,

I’m sorry but I don’t see how you can have diagram with QED>4 for your process.
Could you give me the exact syntax that you use such that i can try to reproduce your diagrams and then give you an appropriate answer.

In the mean time, please take a look at the following tutorial concerning the danger of using $$ and / syntax:
 https://cp3.irmp.ucl.ac.be/projects/madgraph/attachment/wiki/Lund2014/14_09_23_tutomg_Lund.pdf
(slide number 24 and following)

Cheers,

Olivier

On Nov 13, 2014, at 4:26 PM, Niklas Garner <email address hidden> wrote:

> New question #257497 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/257497
>
> Hello,
>
>
> I'm interested in investigating VBS, in particular WW scattering e.g. WW -> h -> WW. When I use
>
> generate p p > h p p $$ w+ w- / z a QCD=0 QED=3, ( h > w+ w- QED = 1)
>
> to generate a sample, everything works fine and I get only the diagrams I am interested in (only 17 diagrams, all with QED = 4). The issue that I am having is that when I change h to either z or a I run into some issues.
>
> I no longer get the same diagrams as before (obviously replacing h with the appropriate boson) but instead a large number (300+) of diagrams mostly with QED = 5+ with the diagrams I am interested in buried somewhere in the mess.
>
> Is there some fix to this, or is there something physically motivated preventing from looking at the diagrams I want?
>
>
> Thanks,
>
> Niklas
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Here is my proc_card:

set loop_optimized_output True
set group_subprocesses Auto
set ignore_six_quark_processes False
set gauge unitary
set complex_mass_scheme False

define q = u c t d s b
define p = g u c d s u~ c~ d~ s~
define j = g u c d s u~ c~ d~ s~
define l+ = e+ mu+
define l- = e- mu-
define vl = ve vm vt
define vl~ = ve~ vm~ vt~

generate p p > h p p $$ w+ w- / z a QCD=0 QED=3, ( h > w+ w- QED = 1)

Hi,

thanks for the lines, but I actually need the case for the photon/Z is this
generate p p > a p p $$ w+ w- / z h QCD=0 QED=3, ( h > w+ w- QED = 1)
generate p p > z p p $$ w+ w- / a h QCD=0 QED=3, ( h > w+ w- QED = 1)

Cheers,

Olivier

Sorry, I thought that might be clear from the original post. For the photon/Z samples I simply replaced h with a or z and replaced the corresponding boson after the / with h.

Does that make sense?

Niklas

Hi Niklas,

> Sorry, I thought that might be clear from the original post.
Actually it was, but this is incompatible with your statement that you get QED=5 diagram. This is just impossible due to the process/syntax that you use. and indeed I do not reproduce the fact that your statement give you QED=5.

> Does that make sense?

Since you use very dangerous syntax, I will not comment on the point if they make sense of not. This is your job to ensure that interference with diagram that you remove are indeed negligible and that such removal still allows MG to computer gauge and lorentz invariant quantity (you can use the command check to check those point).

Now, I have look at the diagram that you allow with your syntax, and I do not see any physical reason to remove any additional diagram. Those diagrams have non-negligible interference (at least I think so) and those should be taken into account. To strenghten this point, even the VBFNLO program which is dedicated for such kind of processes, (and assumes VBF cut to enhandce the VBF region and thererfore discard some diagram) have those diagrams included.

Cheers,

Olivier

I apologize, I was mistaken. They were, in fact, QED = 4. Thanks for the clarification, I think I understand what you are saying.

Two follow up questions:

a) Is there some syntax I can use to restrict to the diagrams which are identical the those I reproduced in the Higgs channel, only replacing the intermediate boson?

b) If so, how can I determine whether they are negligible? Would I check whether the cross section changes by a significant amount upon removal of the diagrams?

Thanks for your patience,

Niklas

Hi,

>a) Is there some syntax I can use to restrict to the diagrams which are identical the those I reproduced in the Higgs channel, only replacing the intermediate boson?

For the photon certainly not (this would be completely crappy for sure) and for the Z, I do not see any that would fit.

>b) If so, how can I determine whether they are negligible? Would I check whether the cross section changes by a significant amount upon removal of the diagrams?

Cross-section is not enough. You need to check shape as well (especially in queue of distribution).
 In top of that, you need to check that gauge invariance is working as well in your full phase-space. Note that your result will not be Lorentz invariant if the amplitude is not gauge invariant (which can happen easily if you remove some important diagram).
In practise, it is enough to check the Lorentz invariance of your matrix-element to ensure that your matrix-element is gauge invariant. (I have only face one case where the matrix element was Lorentz invariant but not gauge invariant).

Cheers,

Olivier