From groer@fnal.gov Tue Dec  9 23:36:28 2003
Date: Tue, 09 Dec 2003 23:32:06 -0600
From: Leslie Groer 
To: d0diem@fnal.gov
Subject: Status Reports from D0 DiEM Subgroup

Status Reports from D0 DiEM Subgroup - Dec 9, 2003
==================================================

From Greg Landsberg
-------------------
I am working essentially on what I have shown at the NP workshop
Will update my sample with new files processed by Alex and will continue
working on optimizing the track isolation cut and studies of the
likelihood and HM7 instead of HM8.

One thing worth mentioning to the diEM subgroup and EM group:
- I found that the standard d0correct EM energy scale does not quite
  reproduce the Z-mass. I have to apply additional correction factor of
  1.007 in CC and 0.995 in EC in order to get it right in all three
  cryostats.

From Oleksey Atramentov
-----------------------
I am working on SMT and CFT hits associated with EM object and there is
a hope that it will improve fakes (see my talk at Calor workshop, also
minutes from fridays dilepton meeting).  I don't have anything
presentable for tomorrow.

From Alex Melnitchouk
---------------------
[from conversations and a bunch of e-mails]
Alex has been hunting down the phi asymmetry in the CS 2EMhighpt sample
See 
http://www-d0.hef.kun.nl///askArchive.php?base=agenda&categ=a03646&id=a03646s1t6/transparencies and

http://www-d0.fnal.gov/~melnit/Dec05_phi_distributions_for_ECemobjects_p14CS2EMhighpt.eps

In this pursuit we (re)discovered that the current certified EMID v5.0
is not being applied in the current tarball of d0correct v00-00-04
available from the Common Sample Group.  There are new energy scale
corrections for each quadrant of the calorimeter (4 regions per
cryostat) based on a sample of Zee events.  There is a significant
difference between the North and South Endcap cryostats and between
top and bottom within each of the Endcaps.  See Smain Kermiche's 
talk at the CALGO workshop or the EMID certification page
There are also geometric corrections from Anne-Marie Magnan for MC


For definiteness, d0correct v00-00-04 contains the following EM packages:
  emreco v00-12-34
  hmreco v00-06-16
  em_util v00-02-82
  em_evt js-31oct2003
  empartfit v00-03-12
These tags are for the p13 certification (v4.1) documented in Junjie Zhu's
D0 note 4171

The previous energy scale from Zee events in p13 was based on a study by
Alexis Cothenet, dividing the CC into 4 regions and treating each EC
as one region (i.e. 6 total instead of the 12 used in EMID v5.0).

There are new tagged versions for the EMID certification v5.0:
  emreco v00-12-36
  hmreco v00-06-17
  em_util v00-02-85
  em_evt v00-15-39
  empartfit v00-03-12
A method to tag and provide these from the CSG is still being worked out,
and should be available in a few days.  Other physics groups are checking
these packages out independently already for running d0correct.

From a separate e-mail discussion about triggers
-------------------------------------------------
Most of the p14 data came in with v12 of the trigger list.
From Greg's NP talk, these are the unprescaled (mostly) triggers
that were used in the p14 datasets (from v11 and v12 of the trigger list).

Single EM: EM_HI, EM_HI_SH, EM_MX, EM_MX_SH,
           EM_HI_2EMFR8, EM_MX_2EMFR8,
           EM_HI_2EM5_SH, EM_HI_2EM5_EMFR8, EM_MX_2EM5_EMFR8,
           E1_SH30, E2_SH30, E3_SH30

DiEM:      2EM_HI, 2EM_HI_SH, 2EM_2MD12, 2EM_2MD12_CEM10,
           E1_2SH8, E1_2L15_SH15, E1_2L20,
           E2_2SH8, E2_2L15_SH15, E2_2L20,
           E3_2SH8, E3_2L15_SH15, E3_2L20

The gain in efficiency from using single EM triggers rather than diEM
triggers for most of the high-pT analyses needs to be understood, but
the suspicion is that it won't make a large difference as we are very
efficient for high-pT diEM objects.  Once you are OR-ing triggers
though, then adding other unprescaled triggers should not cause any
difficulties in calculating luminosities.  If prescaled triggers are
used, then this gets a little more complicated.  Greg argues that the
fake samples will be less biased if we use the OR of single EM and diEM
triggers.  E1,2,3 have all run unprescaled so far as we have not
reached 60E30 in luminosity (this needs to be checked as trigger files
were tuned during this period and prescales sometimes have been modified
if there were anomolous trigger rates).

John Gardner showed preliminary turn on curves for the triggers at the
W/Z meeting today.  He will work on the efficiency for the OR.