MadGraph5_aMC@NLO

sbottom pair production cross section increase with decay

Asked by Stephan Lammel

We noticed that the sbottom pair production cross section changed on us depending on the decay.

Using MadGraph v2.1.2 we get for sbottom pair production of 100 GeV at the LHC a cross section of
333.7 pb. This compares well to 362 pb of Prospino. Adding the sbottom decay into b-quark and LSP
increases the reported production cross section to, for instance 9025 pb for m_LSP=90 GeV. We (or
MadGraph) are obviously doing something wrong.

Reading through previous cross section questions, https://answers.launchpad.net/mg5amcnlo/+faq/2442,
we don't see those reasons applying. All should be small or reductions in cross section but not increase it
by a factor of 27 (but we removed final state cuts and also changed BWcutoff).
The masses of all other SUSY particles are set to 10 TeV (it seems to make no difference), the sbottom
decay into b + neutralino_1 is set to 100% branching ratio, and the decay width kept at 300 MeV from the
original SLHA file. Is MadGraph may be recalculating width/partial width from other model parameters in
the SLHA file? (Is there a way to avoid specifying a full model but provide just the relevant parameters for
the process?)

Any idea what could be wrong?
Thanks,
 - Stephan

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 Stephan,

MadGraph is never using the information of Branching ratio.
The cross-section depends on the value of the total width and of the coupling.
Approximatively, you can see it as if MG is recomputing the partial width but not the total width.
Since you keep the total width unchanged for your computation, this is likely to be the reason of the error.

My advice would be to set the width information to Auto.
(i.e. having the line like “DECAY 100024 Auto” in your param_card)
MadGraph will then call the new MadWidth package (arXiv:1402.1178) to compute the total/partial width for that particle.
The card is then replaced by a new one with this information before the computation of the cross-section.

Cheers,

Olivier

On Aug 25, 2014, at 11:37 PM, Stephan Lammel <email address hidden> wrote:

> New question #253601 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/253601
>
> We noticed that the sbottom pair production cross section changed on us depending on the decay.
>
> Using MadGraph v2.1.2 we get for sbottom pair production of 100 GeV at the LHC a cross section of
> 333.7 pb. This compares well to 362 pb of Prospino. Adding the sbottom decay into b-quark and LSP
> increases the reported production cross section to, for instance 9025 pb for m_LSP=90 GeV. We (or
> MadGraph) are obviously doing something wrong.
>
> Reading through previous cross section questions, https://answers.launchpad.net/mg5amcnlo/+faq/2442,
> we don't see those reasons applying. All should be small or reductions in cross section but not increase it
> by a factor of 27 (but we removed final state cuts and also changed BWcutoff).
> The masses of all other SUSY particles are set to 10 TeV (it seems to make no difference), the sbottom
> decay into b + neutralino_1 is set to 100% branching ratio, and the decay width kept at 300 MeV from the
> original SLHA file. Is MadGraph may be recalculating width/partial width from other model parameters in
> the SLHA file? (Is there a way to avoid specifying a full model but provide just the relevant parameters for
> the process?)
>
> Any idea what could be wrong?
> Thanks,
> - Stephan
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Stephan Lammel (lammel) said : 		#2

Thanks Olivier! Recalculating the partial width certainly explains it.
(The MadGraph calculated partial width is a factor 5.2 larger than the
old total width and 5.2 squared, for sbottom and anti-sbottom, is 27.)

If MadGraph re-calculates all the branching ratios, it would be nice
to log discrepancies or save the used values not only in case of an
"Auto" in the decay table. (Just a thought.)

Thanks,
 - Stephan

Revision history for this message
Stephan Lammel (lammel) said : 		#3

Thanks Olivier Mattelaer, that solved my question.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#4

Hi,

> If MadGraph re-calculates all the branching ratios, it would be nice
> to log discrepancies or save the used values not only in case of an
> "Auto" in the decay table. (Just a thought.)

Actually this is not possible since we do not compute those number. That’s why I said “approximatively” in my previous email.
This is not what we compute but what we compute is actually equivalent (to gamma/M accuracy) to the partial width (if you know the production cross-section that we do not compute either)

Cheers,

Olivier

On Aug 26, 2014, at 6:26 PM, Stephan Lammel <email address hidden> wrote:

> Question #253601 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/253601
>
> Stephan Lammel posted a new comment:
> Thanks Olivier! Recalculating the partial width certainly explains it.
> (The MadGraph calculated partial width is a factor 5.2 larger than the
> old total width and 5.2 squared, for sbottom and anti-sbottom, is 27.)
>
> If MadGraph re-calculates all the branching ratios, it would be nice
> to log discrepancies or save the used values not only in case of an
> "Auto" in the decay table. (Just a thought.)
>
> Thanks,
> - Stephan
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.