Kaon_LT Experiment

                 Sept 12/24 PionLT/KaonLT Analysis Meeting Notes

                 -----------------------------------------------

                                (Notes by GH)

                     Today: KaonLT 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, Nathan Heinrich, Alicia Postuma, Vijay Kumar,

   Nacer Hamdi, Muhammad Junaid, Ali Usman

Ohio - Julie Roche

JLab - Dave Gaskell

Virginia - Richard Trotta

CUA - Tanja Horn, Casey Morean

FIU - Pete Markowitz

Alicia

------

pi+ BSA paper Referee comments

- general interest criterion:

  - need to modify title (both editor & referee A)

  - some parts too technical, reword or move definition earlier

  - Tanja: stress importance of confirming CLAS measurements w/ different

    systematics

- improve discussion of theory vs data agreement

  - Tanja: change from discussion of models, to discussion of the physics

    represented by the models

  - *NB* with improved data and models, we can make a more definitive statement

    than prior works on the approach to the GPD regime

    - GPD regime not reached yet, as the GPD model does not describe data well

    - interestingly, there is a non-perturbative model that encapsulates the

      major features of the data

- modify figure to make syst uncertainties more clear

  - Dave: *NB* remove cap on outer error bar

  - Tanja: if new figure is not an improvement, include it in a note to the

    referee, so they're aware we tried

- comment on backgrounds

  - pion purity - side study using HGC w/hole cut, detector efficiencies

- overall, fairly optimistic on eventual acceptance

- expecting a new draft for comment in 1-2 weeks

Richard

-------

KaonLT LT-separations

- plot of Data/MC ratio per t-phi bin for high, low epsilon data combining all

  settings

  - ratios are 0.1-0.5 with some variation

- will try reproducing Vijay's data using RT's code, hoping to find what the

  problem is

  - has run# for center setting from Vijay, replaying data now

Vijay

-----

Next steps for low Q2 pi+ LT-separations

- setting up SIMC for ALMA9

- first step is to understand why HMS detection volume cut dependence is so

  large

- then Q2=0.42 data

- Fpi extraction will wait until other studies done (proton absorption,

  systematics)

- shows LT-separations from thesis

  - had changed LT,TT functions from last update

  - sigLT increased further at lowest t-bin when changing bin

  - ratios flatter than last update, particularly at low -t

  - *NB* ratios still have some oscillations for higher -t bins, indicates that

    t-dependence in LT,TT functions still not right

Ali

---

piDelta BSA analysis

- shows new MM plots per t-bin, Q2=2.1, W=2.95

  - plots are for thesis, asking for feedback

  - Delta shape comparison: (Data-Bkd)=RED,  Delta MC=BLUE

    - agreement is pretty good in main peak region, but divergence is larger on

      edges

    - Red is integrated for BSA, not the Blue

    - Garth: should make sure integration region is where Red and Blue have

      good agreement

      - MM cut values: 1.11-1.40

      - *NB* need to overlay these on the plots to check this more clearly

Nacer

-----

KaonLT Heep analysis

- Data/MC ratios too large after applying proton absorption correction

- updated John Matter's materials table for proton abs calculation

  - removed NGC

  - adapt values for actual aerogel thickness

  - obtain absorption factors:

    3.8+4.9    n=1.030  7.02%

    6.2+10.6   n=1.015  6.30%

    8.2        n=1.011  6.09%

- review of proton correction method in W.Li's thesis (originally developed by

  Henk Blok)

  - plot of MMp @ 3.8 GeV

    - data and SIMC agree up to 0.02

    - data tail grows past 0.02, disagreement w/MC degrades

    - MC tail slope gets steeper past 0.04, while data is more steady

    - analyzing with tigher MMp<0.02 cut gives 5% lower Data/MC ratio

    - *NB* Dave: it's important to choose one cut and apply it.  Keep the

      analysis simple.  Don't apply both EMiss and MM cuts at same time

  - shows SHMS-beta plot for Heep events & compare to Bill's data

    - Dave: beta trigger is set up differently now, pretty sure you won't get

      the tail as in 6 GeV era

    - Dave: *NB* SHMS-beta distribution looks weird, need to check variable

      beta-p for corrected variation in TOP across focal plane

  - tries replicating Bill's proton absorption correction method

    - get fraction ~95%, so no big change from full factor in J.Matter thesis

    - Bill did have a tigher MM cut than Nacer

- next steps

  - Dave: *NB* suggests to simplify the situation by looking at single arm Heep

    to see if Data/MC ratio is 1 or not

    - if Heep singles OK, it indicates the problem is with the proton arm in

      Coin

    - if Heep singles NOT OK, problem is with the electron arm or beam

      normalization

Junaid

------

PionLT Heep analysis

- Data/MC ratios w/ and w/o MM cut, to eliminate pi0 contamination

- HMS matrix element check

  - discovered that param files using "standard" HMS ME for ALL settings, no

    special ME being used for P_HMS>5.5 GeV/c

  - tried using KaonLT ME for P_HMS+6.59 for 10.5 GeV data

    - subtle differences from standard ME, slightly better Data-MC agreement,

      but nothing dramatic

  - Dave: a student is working on P_HMS=5.78 GeV/c MC

  - Julie: will check who is working on ME from NPS data

  - Tanja: *NB* take a look at the slides at:

    https://indico.jlab.org/event/866/contributions/14919/subcontributions/289/attachments/11532/17851/Ploen_CollabOpticsCalibNPSJuly2024.pptx

    - slide 12 gives a useful list of who is working on which ME

    - P=5.639, Julio Gil Guiterrez & Abhyuday Sharda

    - P=5.878, Josh/Christine

    - P=6.117, Josh Crafts

    - P=6.667, Christine Ploen

- issues w/ Hodo-3/4 effs

  - looking at hcana code, tried removing PID cuts, got same number

  - Dave: *NB* for your preliminary Heep yields, try using a high efficiency

    for now, we will eventually get the effs high as expected

    - should be ~99% except for runs w/ low momentum in quartz plane

  - MJ: *NB* we should be able to fix Heep ratios if we solve problem with 3/4

    trig effs

    - Nacer also sees too-low 3/4 effs in KaonLT data, so it seems there's a

      common cause for both analyses, the Heep singles analysis that Dave

      suggests will make this more clear

Nathan

------

PionLT luminosity studies

- preparations to look at 2-arm daq ratios from SIDIS lumi scan

- modifying analysis sript, not there yet

Next Meeting

-------------

- Thur Sept 19 @ 16:30 Eastern/14:30 Regina

  - PionLT will go first