Oct 3/24 PionLT/KaonLT Analysis Meeting Notes
                 ---------------------------------------------
                                (Notes by GH)

                     Today: PionLT will be discussed first

Please remember to post your slides at:
https://redmine.jlab.org/projects/kltexp/wiki/Kaon_LT_Meetings

Present
-------
Regina - Garth Huber, Muhammad Junaid, Nathan Heinrich, Vijay Kumar,
   Nacer Hamdi, Alicia Postuma
FIU - Pete Markowitz
Ohio - Julie Roche
Virginia - Richard Trotta
CUA - Tanja Horn
JLab - Dave Gaskell

Nathan
------
PionLT COIN Luminosity studies
- made a mistake in setting up scripts, had e+ tracking eff instead of proton
  in SHMS
- replaying files again, hopefully will have some results for next week

Junaid
------
PionLT PID studies
- completed HMS e- selection
- now doing SHMS pi+ selection
  - using 9.177 GeV data
  - using RF time, Aerogel, HGC to select clean sample
  - will look at 3 physics settings, one where SHMS is at 5.5deg to check for
    backgrounds
- planning to have a report next week

Richard
-------
KaonLT LT-separations
- looked at W-dependence, comparing Q2=3.0, W=3.14 with Q2=3.0, W=2.32
- adjusted exponent in Wfac=1/(W^2-mp^2)^n from n=2.0 to n=2.25 and n=3.40
  - Tanja's global K+ analysis had n=2.25
- seems to put ratios near 1 instead of 0.3-0.5
  - Dave: confused about this.  Changing the exponent should only afffect the
    Data/MC ratio for 1st iteration, but ratios would get rescaled in further
    iterations
  - Richard: the ratio seems to persist near 1 for first few iterations, will
    try more
- will look at both Q2=3.0 settings for comparable t-ranges and see what is
  best value of exponent
  - would try different n manually, not as a free parameter in the fit
  - then would look at other Q2 settings, to see if a global n value works or
    not

Vijay
-----
Low Q2 PionLT systematic uncertainties
- earlier studies showed HMS delta was very sensitive to +/-10% change in cut
  value (as in Tanja's thesis)
  - Dave: should not go past +/-8% HMS delta range, as Matrix Elements are not
    as well constrained.  In general, for both spectrometers keep within the
    nominal cuts and only make them tighter, not looser
- found that only the low epsilon data covers +/-8% delta range, mid and high
  epsilon data do not fill whole focal plane, due to the applied diamond cut
  (obvious in retrospect) and are not affected by the +/-10% cut variations
  - doing an LT-separation with this cut will not make sense, as the cut
    variation effectively makes the diamond smaller for low epsilon data, but
    not the mid or high epsilon data
- will compute sig_UNS instead, for only low epsilon data
  - this will be more comparable to the study shown in Marco's thesis

Data replay and preservation
- planning to replay Q2=0.42 data next week
- unfortunately, the Q2=0.38 data were not backed up, will replay it as well
- Tanja: it is important to also archive the code used to generate the data, so
  they're linked together
- Tanja: it is also important to do some tests to make sure the new Q2=0.38
  replay is not different than the old one

Alicia
------
BSA manuscript changes
- please review the draft manuscript on OverLeaf
- went through draft response to editor and Referee B
  - incorporating GH's comments then will circulate to group via PDF
- would like to resubmit to PRL by Oct 11

Nacer
-----
KaonLT Heep singles
- finished replays
  - need to set up config files for report files
- Junaid sent SIMC configuration files for singles


Next Meeting
-------------
- Thur Oct 10 @ 16:30 Eastern/14:30 Regina
  - KaonLT will go first