Tau identification in a hadron collider environment is not an easy task, yet many theories predict copious decay of new partices into taus; therefore, tau identification may very well be the key for physics discovery in Run II at D0. Find complementary information at the D0 Tau ID group's homepage.
switch to my:
| PMCS Tau page | A parameterized fast Monte Carlo for taus. |
| L2 Tau page | A level 2 global tau trigger. |
Information which follows:
back to: Bryan's homepage
The people who are currently working in the tau ID group:
| Name | |
|---|---|
| Qizhong Li | qzli@fnal.gov |
| Paul Padley | padley@fnald0.fnal.gov |
| Gustaaf Brooijmans | gusbroo@fnal.gov |
| Bryan Smith | brsmith@fnal.gov |
| Dhiman Chakraborty | dhiman@fnal.gov |
| Kwok Chan (a.k.a. Leo) | klchan@fnal.gov |
| Yuri Gerstein | gerstein@fnald0.fnal.gov |
| Naresh Sen | nareshs@ruf.rice.edu |
| Silke Duensing | duensing@fnal.gov |
back to the top
Here is a W to Nu Tau in which the Tau has a 1 prong pion decay. I show two views: the 3D "mrd" view of the whole detector and the lego view of the calorimeter.
One must add the "event_display" package to get the 3D view and the "legoplot" package to get the lego view (hopefully, all of this stuff gets incorporated into one package in the near future).
| Package | RCP | Exec | View |
|---|---|---|---|
| event_display | event_display.rcp | mrd | 3D |
| legoplot | LegoPlot.rcp | legoplot_x | lego |
Here is an example of what you might do from your release area:
back to the top
SAM is a D0 project that is supposed to help physicists store and access the large Run II datasets. It is in a state of flux as problems are ironed out, so be prepared for problems.
To get your analysis code running, try the quick start guide.
Before doing anything, try a few commands to make sure SAM works, for example:
setup sam
sam help
sam locate file (to see if alive or not)
sam dump fss (should spew a bunch of junk)
sam dump station --projects Search the SAM database
with the string
Type the following two lines to start the application that will
allow you to "make a SAM project", from d0mino:
Once the project is defined and saved, don't forget to make a snapshot of what you are doing.
Read the SAM Mailing List archive before complaining to sam-users@fnal.gov.
Note: if you want to know what is "pinned" on the SAM disks, look
in /sam/cache/* and you can find out. (There is
probably a better way to do this.) In an emergency, like SAM being
really dead, one can just run on these pinned files directly.
A list of simulated "mcc99_2" events that are available in SAM is maintained by Greg.
back to the top
For you reference, below is a concise summary of the Run I W to Tau Nu analysis at D0, which consisted of a measurement of the W to tau nu cross section times branching ratio (now published in Phys. Rev. Lett. 84, 5710 (2000).
down to:
It must be remembered that D0 will, to a large extent, be a completely new detector in a much higher luminosity environment for Run II, hence the tau ID strategy will change, primarily by incorporating the tracking information.
back to the top
First, here is a sketch of the Run I Tau trigger:
Remember, there was no magnetic field at D0 in Run I, which limited tracking trigger capabilities. The Run II tau trigger will be different, combining tracking and calorimeter information.
back to the list
In the offline analysis, here is a list of the Run I the selection requirements for W --> Tau Nu --> Hadrons:
Questions about the previous cuts:
It would have been nice to see of plot of what the affect of each cut was. Back of the envelope, some of the cuts seem to have drastic overlap. I was able to find a copy of Hailin Li's thesis, referenced in the paper; hoping to glean information from it.
back to the list
Obviously, multijet events with energy fluctuations that mimic MET are a serious source of background. Here are a list of backgrounds:
| Source | Contamination |
|---|---|
| QCD | 9% |
| Electronic noise | 7% |
| Z -> Tau Tau | 3% |
| W -> Electron Nu | 0.2% |
QCD events and electronic noise were each background contaminations approaching 10% of the number of observed events.
The QCD background was determined from the data by taking ratios based on the profile variable in "signal" and "background" regions with and without MET cuts applied. It is claimed that the profile and MET are not correlated. Clearly, they are not completely uncorrelated for the taus are part of the sample! Personally, I would have liked to know what the QCD MC says the background is; perhaps because of the hot cell problem, the QCD MC was not reliable enough to be trusted. In my opinion, the QCD background calculation needs to be redone! Maybe after I read Hailin Li's thesis, I'll change my mind.
In Run II, I hope the electronic noise can somehow be beaten down, yet it was accounted for in a clever way in Run I; see the next section.
back to the list
In the Run I analysis, tau identification is sensitive to the underlying event and to electronic noise in the calorimeter. Cleverly, W-> Electron - Nu data events were used to obtain realistic underlying event and noise in the detector. The electron in the data event was replaced with a simulated and reconstructed kinematically equivalent Monte Carlo tau that was decayed hadronically in the detector.
back to the list
The W to Tau Nu cross section was extracted in the obvious way. No disagreement with the Standard Model found.
Run II will allow a much improved measurement of this cross section; hopefully, the tracking can be used to make a much better estimation of the QCD background. But most importantly, other measurements involving the tau will also be possible.
back to the list
Send comments and suggestion to brsmith@fnal.gov
Last modified: Aug. 17, 2000