[Rivet] Jets in DIS

[Roman Kogler]

Dear Rivet authors and developers,

I agree with Hannes. On the experimental side we measure the electron 
with different Pt and energy criteria depending on the scattering polar 
angle theta (at least in H1, since we have two different detectors for 
the electron measurement, depending on theta). However, we know the 
efficiencies of the electron finders very well and correct for these. 
The measurements are then given in a region of phase space given in 
terms of Q2, x and y, always assuming perfect electron identification. 
In Rivet, we do not want to introduce efficiencies of the electron 
finding and thus deplete regions of phase space due to events lost.

On the other hand, Jon's statement is of course correct. An electron 
identification based on what the exeriments did would be preferable, 
since there is an ambiguity in multi-lepton final states. I like the 
suggestion of the scattered electron to be the highest energy lepton in 
a given pseudorapidity region. We should be careful though to define the 
measureable region in analogy to the acceptance of the experiments. I 
believe that the backward region covered by H1 and ZEUS is theta<176 
degree, which is roughly eta=-3.35. So the eta region for the electron 
finding should read something like -3.5 < eta < 4.

One thing to think about, though: In the final states of the present MC 
event generators for multilepton production in DIS, the leptons from the 
MC matrix elements are treated differently than leptons from particle 
decays. Whereas in the first case the outgoing leptons are calculated 
taking interference effects into account, leptons from particle decays 
are modeled by fragmentation functions which do not follow the rules of 
quantum mechanics (you could say that only the real part of the 
amplitude is considered). So we actually know where leptons from 
particle decays originate from. I think that finding the scattered 
electron before the hadronisation starts would be the way to go. Since 
this is not accessible experimentally, an isolation criterium as Jon 
suggested will be good to have.

Cheers,
    Roman

P.S. I know that there are only a few HERA papers in Rivet so far, but I 
hope that this will change soon. I'm working on it and I know at least 
from one more analysis from H1 which will publish a paper together with 
a Rivet analysis soon.




On 10/19/11 9:43, Jonathan Butterworth wrote:
> Hmmm.
>
> Except that in principle a "scattered" lepton is only identifiable as 
> such via its final state characteristics. In a multilepton final state 
> no one tells us which it is by any other means. That's quantum mechanics.
>
> The "generator independent" way Hannes mentions was, I think, just a 
> bunch of if/then statements looking at the MC matrix element: not 
> really allowed.
>
> TBH this is something that has worried me on HERA results since I 
> started worrying about such things (about halfway through being ZEUS 
> photoproduction convener I think).
>
> In practice I think choosing the highest energy lepton with
> 4 > eta > -2.5 or something is what the experiments did (though at 
> detector level) and should work for rivet. An isolation requirement 
> might also be relevant I guess. I believe for nearly all analyses any 
> misID is a tiny effect, and since the experiments generally didn't 
> measure the scattered lepton unambiguously via a final state 
> definition, it's probably the best we can do.
>
> I'll have a read of some old papers and try and check this.
>
> Cheers,
> Jon
>
>
>
> On 19/10/2011 18:01, Hannes Jung wrote:
>> Hi Leif, Andy et al
>>
>> the measurements are in terms of Q2,y,x, so the definiton of the
>> scattered electron does no longer enter here. In hztool we had a
>> definiton of the scattered electron idependent on the generator,
>> and it stored the scattered electron.
>> Now since the effect of misidentification of the scattered electron is
>> already included in the correction of the data, one needs a defintion,
>> which really identifies the scattered electron.
>> There should be not different ways to identify the electron, there is
>> only ONE scattered electron in DIS.
>>
>> Cheers
>>
>> Hannes
>>
>> On 19 Oct 2011, at 17:40, Andy Buckley wrote:
>>
>>> On 19/10/11 12:24, Leif Lönnblad wrote:
>>>> On 2011-10-19 10:04, Hannes Jung wrote:
>>>>> hm, taking just the frist electron might be tricky, as it is not 
>>>>> based
>>>>> on physics.
>>>>
>>>> I'm a bit surprised about this discussion. The whole point with 
>>>> rivet is
>>>> to do what the experiments do. H1 had one (or maybe several) very
>>>> specific way of finding the scattered electron. For an H1 analysis, 
>>>> this
>>>> should be reproduced in Rivet. The modular structure of Rivet means 
>>>> that
>>>> we can implement several electron finders (subclasses of DISLepton). A
>>>> given rivet analysis should use the one which corresponds to the one
>>>> used in the experimental analysis. And if the ones included in the 
>>>> rivet
>>>> distribution does not fit a specific analysis, one would have to
>>>> implement a new one. Or am I misunderstanding the discussion?
>>>
>>> I don't think you're misunderstanding the discussion, no! So if the
>>> HERA people on this email can point us to the scattered lepton
>>> experimental definitions for the few HERA papers implemented in Rivet,
>>> we'll happily implement them as appropriate.
>>>
>>> Cheers,
>>> Andy
>>>
>>> -- 
>>> Dr Andy Buckley
>>> SUPA Advanced Research Fellow
>>> Particle Physics Experiment Group, University of Edinburgh
>>>
>>> The University of Edinburgh is a charitable body, registered in
>>> Scotland, with registration number SC005336.
>>>
>>
>>
>> ***********************************************************************
>> Hannes Jung
>> Email: Hannes.Jung at cern.ch <mailto:Hannes.Jung at cern.ch>
>> mobile :+49 40 8998 93741
>> http://www.desy.de/~jung
>> Tel: +49 (0) 40 8998 3741 (DESY)
>> Tel: +41 22 76 62602 (CERN)
>> CERN - PH
>> 42-2-033
>> CH-1211 Genève 23
>> Switzerland
>> ***********************************************************************
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> Rivet mailing list
>> Rivet at projects.hepforge.org
>> http://www.hepforge.org/lists/listinfo/rivet
>