Diphoton bruduction 2_2__j__j__P__P freezes

[strebeli@ethz.ch]

Dear Sherpa users,

the diphoton production freezes if a pt cut on the diphoton transverse momentum is applied. After the message "lean back and enjoy" nothing happens:

Grid_Creator::CreateGrid(0x216e8fc0): Initializing grid for MI.

99 % ( 233 s elapsed / 2 s left / 235 s total )

} Simple_Chain::CalculateTotal(): Result is {

\sigma_{hard} = 152.876 mb at p_\perp_{min} = 4.49923 GeV

} Profile_Function_Base::CalculateOMean(8.18693): Results are {

k = 57.9487 <\tilde{O}> = 0.141279

} Initialized the Multiple_Interactions_Handler (MI_Handler). Process_Group::CalculateTotalXSec(): Calculate xs for '2_2jjPP' (Comix) Starting the calculation. Lean back and enjoy ... .

---

(run){ RANDOM_SEED=1; EVENTS=10; }(run)

(beam){ BEAM_1 = 2212; BEAM_ENERGY_1 = 7000.; BEAM_2 = 2212; BEAM_ENERGY_2 = 7000.; }(beam)

(processes){ Process 93 93 -> 22 22 93{2}; Order_EW 2; CKKW sqr(20./E_CMS); End process; }(processes)

(selector){ "Calc(PPerp(p[0]+p[1])<E_CMS&&PPerp(p[0]+p[1])>200.0)" 22,22 1,1 }(selector)

(me){ EVENT_GENERATION_MODE = Unweighted }(me)

(isr){

PDF_LIBRARY MSTW08Sherpa; PDF_SET=mstw2008lo;

}(isr)

(mi){ MI_HANDLER = Amisic #Amisic # None or Amisic }(mi)

[marek]

Hi,

the problem your are encountering is due to a misconception: In your core process, pp->PP, the diphoton pair does not have any transverse momentum and thus never passes your selector requirements. For this, it needs at least one more parton to be present to recoil against. Thus, the lowest order process contributing is pp->PPj. Replacing the processes section with

Process 93 93 -> 22 22 93 93{1}; Order_EW 2; CKKW sqr(20./E_CMS); End process;

should gives you a working setup. However, as the diphoton systems may again additional transverse momentum due to parton shower emissions at a later stage please be aware that the ME-cut needs to be less restrictive than the analysis cut by a good measure.

Another way forward would be generating the sample inclusively (no pT-cut on the diphoton system) nonetheless and combating the ensuing inefficiency by statistically enhancing the region of interest by adding

Enhance_Observable VAR{log10(PPerp(p[2]+p[3]))}|1|3 {3,4,5}

to the process setup. This will generate the events such that the statistical error in the log10 of the transverse momentum of the diphoton system in between the values 1 and 3 is flat, adjusting event weights appropriately. This inclusive ansatz will safeguard you against missing contributions. Please note that it is only applied to the 3,4,5 parton final state matrix element as, again, it is ill-defined on the pp->PP ME.

Please also note in either case: When calculating photon production in the presence of additional jets you also have to safe-guard against collinear q->qP divergences. To this end, put

CSS_EW_MODE=1; ME_QED=0;

into the (run){..}(run) section as this tells the code to include the QED splitting functions and thus regulate these QED divergences in the multijet merging. See also the relevant diphoton examples in the Manual.

Hope this helps, Marek

[strebeli@ethz.ch]

Hi Marek,

I thank you a lot for the fast response and the detailed explanation. I changed my Run.dat according to your description above and everything works fine now.

Cheers,

Ivo