[Rivet] Stable particle lifetimes

[Peter Skands]

Hi guys,

While it seems relatively standard to use c*tau > 10mm at hadron 
colliders (as is also mentioned in the Rivet manual under the analyses), 
it was so far not completely clear to me what stable-particle definition 
to use for the LEP measurements, and I think you have been in doubt too 
- at least the Rivet manual does not give any explicit recipe.

As you know, this can impact multiplicity distributions and also 
momentum fraction distributions (an undecayed particle with high x vs 
two or more lower-x daughters), which we rely on for tuning 
hadronization parameters.

I just wanted to let you know that at least in one LEP paper I've been 
using (one not in Rivet so far), I now actually found the exact 
definition they use. Of course, that's only really for sure for *that* 
paper, but it's better than nothing, and the answer was a bit surprising 
to me.

The paper is L3, http://arxiv.org/abs/hep-ex/0406049v1

On p.26 where they discuss hadron multiplicities and x spectra, they say

"In this correction procedure, we assume all particles with mean 
lifetime greater than 3.3 × 10−10 s to be stable."

Now, if I am multiplying correctly by the speed of light, that 
translates to 10cm, not 10mm!

I briefly spoke to Leif about this, and he figures that could be 
consistent with people in those days basically only setting the pi0 
stable (and K0Long, I guess). He also mentioned that, in principle, 
Rivet could be able to at least check the settings used for the run 
(e.g., by seeing if there are any stable K0S in the actual event 
records) and give an error message if it looks like the user put some 
particle stable that he wasn't supposed to?

Anton and Stefan: this may mean that we want to be running the 
generators with a cut at 100mm instead of 10mm when we do LEP analyses. 
Rivet/Stefan, I'd also like us to look at possibly including this L3 
analysis in Rivet. It is extremely useful since it uses b-tagging to 
separate out light-flavor fragmentation from b-fragmentation, making it 
possible to tune those separately.

As a side remark, the paper also mentions explicitly that all QED shower 
effects have been corrected for, both ISR and also FSR. That's something 
we should put in a comment in the manual if we include that analysis in 
Rivet.

Anyway, just wanted to let you know, and ask if you have learned 
anything more about the definitions used for the analyses available in 
Rivet so far?

For reference, I am attaching plots illustrating the x and Nch spectra 
for Pythia 8 with 10mm and 100mm, respectively. As you can see, there is 
a noticeable affect at high x in the fragmentation spectrum, and there 
is a HUGE effect on the Nch distribution.

Cheers,
Peter
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Pythia-100mm-Nch.eps
Type: image/x-eps
Size: 17334 bytes
Desc: not available
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20110202/1dbfbe42/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Pythia-100mm-x.eps
Type: image/x-eps
Size: 19542 bytes
Desc: not available
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20110202/1dbfbe42/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Pythia-10mm-Nch.eps
Type: image/x-eps
Size: 16584 bytes
Desc: not available
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20110202/1dbfbe42/attachment-0002.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Pythia-10mm-x.eps
Type: image/x-eps
Size: 19707 bytes
Desc: not available
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20110202/1dbfbe42/attachment-0003.bin>