Kaon_LT Experiment

                 May 29/25 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, Nathan Heinrich, Ivan Zhenchuk, Alicia Postuma,

   Muhammad Junaid, Nacer Hamdi, Vijay Kumar

FIU - Pete Markowitz

Virginia - Richard Trotta

JLab - Dave Gaskell

JMU - Ioana Niculescu

Ohio - Julie Roche

CUA - Tanja Horn

Nathan

------

PionLT deadtime effects

- follow-up to weird dip in HMS-delta (H.gtr.dp vs H.gtr.beta) found by Dave

  Gaskell, run 14613

  - also seen by Junaid in HMS-delta comparison to SIMC

  - Dave says the dip only appears if he places a narrow cut on the prompt CT

    peak

  - Nathan reran Hodoscope calibrations, doesn't see any dip, no change before

    and after calib

    - will look again at Reference Times

- return to this topic near end of meeting

  - Dave shows a plot of CoinTime vs position at 2X plane in HMS, run 14613

  - 2 clear shifts in time are seen, associated with specific paddles

  - suggests that Nathan look at the param files to see what might be wrong

- Carbon Lumi scans after Reference Time fixes

  - Tracked yield is fairly flat, looks generally acceptable

    - however, there are a few outliers.  These are being looked at, either to

      fix them, or find a reason to exclude them

  - before the Ref Time fix, Carbon was flat, so it should be possible to get

    it flat again

- LH2 boiling scans after Reference Time fixes

  - scaler analysis flat (with old boiling corr applied)

  - NoTrack and Track analyses not flat

    - still investigating, no details yet

Junaid

------

PionLT LTsep initial setup

- last week showed calculated Data/MC ratios, where SIMC uses the default

  Bosted physics_pion.f model

  - still going through the Fortran codes

  - next step will be un-iterated sig_unsep, using Bosted SIMC model

- will show draft CAP slides next week

Ali

---

piDelta BSA

- GH gives a few remarks from Ali, his thesis defense is tomorrow

  - asymmetry extraction from rebinned data is done

  - LT'/sig0 ratios with new binning also done, results look promising

  - will show results and draft CAP slides next week

Nacer

-----

KaonLT LTsep initial setup

- shows calculated Data/MC ratios, where SIMC uses the default Koltenuk

  physics_kaon.f model

  - problematic missing bins resolved, the problem was caused by low statistics

    phi bins for Left and Right SHMS settings, resolution was to set the errors

    for these bins to be very large when calculating the average ratios

 - ratios have nearly symmetric phi dependence, which is good

   - some notable phi-oscillations, we note that the Koltenuk model has no

     LT,TT terms, hence no phi-dependence, which we will resolve when doing

     iterations

- un-iterated sig_unsep, using Koltenuk SIMC model

  - calc_xsect.f gives a theta* disagreement between data and calc, will have a

    short discussion with Richard about this at end of meeting

  - LTsep framework is mostly set up now

  - will show draft CAP slides next week

Vijay

-----

pion and proton "absorption" study for Q2=0.425 PionLT data

- pi+, P_SHMS=2.326 GeV/c

  - no Cherenkov cuts uses in analysis, only CoinTime PID needed to get clean

    data

  - Geant4 gives 2.16% bad events, so correction factor=0.9784

- Heep protons:

  P_SHMS=1.729  bad=3.43%

  P_SHMS=2.300  bad=3.50%

  P_SHMS=2.792  bad=2.84%

  - 10k events run for all 3 simulations, finds the higher middle bad value a

    bit odd

  - GH: suggests to run at least 50k events for all settings (including pi+) to

    see if the Heep number changes

  - Alicia: if the middle absorption value says higher with more stats, it

    means this is due to the Geant4 cross section momentum dependence

    - absorption is expected to be higher for P<1 GeV/c, should be fairly flat

      for higher momenta

- will implement absorption correction into physics yield calculation

- Vijay's contract ends in ~10 days, his future plans still uncertain

Richard

-------

KaonLT high Q2 LTsep

- fixed very high theta* values problem noted earlier

  - issue was that the recoil mass in eps_n_theta.f was still set to proton

    instead of lambda mass

  - the theta* values are now lower, initial results show an improvement in

    higher t-bin ratios

  - unfortunately, the change did not get properly pushed to his main analysis,

    rerunning the data again

Ivan

----

Low Q2 KaonLT BSA

- K+ PID investigations

  - switched from HGC PID (shown last week) to RF PID that Nacer uses

  - RF time is not properly aligned at zero for Kaons, unlike Nacer's analysis

    - Nacer had adjusted RF time but the updated Hodoscope calibs did not

      properly get pushed to GitHub

    - Alicia will coordinate with Nacer on a new replay using Nacer's calibs

      and Alicia's helicity scripts

- K+ MM spectra with dummy target and pion leakthrough subtractions

  - the pion leakthrough subtraction seems to work well, leaves clean Lambda

    and Sigma peaks on little background

    - a future study will have to be to vary the pion subtraction and see how

      the BSA results change

- first look at K+ BSA results

  - integrates around the clean Lambda peak, only ~1/3 of Center SHMS setting,

    no Right and Left SHMS yet

  - errors look large, but the beam polarization is only 38%, which magnifies

    uncertainties by ~2.6x

  - mostly sees random scatter around zero

  - we note that YCK model (for pi+n) predicts small BSA at low Q2, the

    asymmetry grows to a plateau value above a Q2-threshold

- also looks at pi+ BSA for same data

  - the pion leakthrough sample is used for this check, calculated the same way

    as for K+

  - asymmetry is about +/-0.02, so very small

  - a clear sinusoidal dependence is seen, not the random scatter as for K+

- also looks at BSA of pi+ background underneath the Lambda (corresponding to

  piDelta region)

  - the asymmetry is very small, less than the +/-0.02 seen for the exclusive

    pi+n

  - this is an important systematic check on the analysis that we will need for

    any publication

Alicia

------

Draft CAP slides on u-channel motivation and analysis

- Bill Li had a recent talk using the Verton (baryon junction) motivation, but

  that is more closely linked to u-channel SIDIS than u-channel exclusive

- also shows Zach Sweger talk from CFNS Baryon Junction workshop, where he

  frames the motivation in terms of TDA and distribution of Baryon # vs

  Momentum

- also shows CLAS-6 BSA results for pi+n channel, comparing forward (t-channel)

  and backward (u-channel) and the unexpected sign flip in the BSA between them

- Tanja comments:

  - one of the issues is that it's hard to find clean Verton observables, to

    avoid a lot of questions that may be difficult to answer, recommends not to

    make it the main emphasis of the talk

  - suggests to emphasize the data and what kind of impact they are likely to

    make

  - it's okay to mention Verton, but don't over promise

  - suggests to focus more on TDAs and use Verton as one way to describe why

    TDAs may be interesting

- Garth:

  - given the audience, comparing more clearly the forward and backward angle

    regions would really be helpful, not only in terms of the differences of

    the detected processes, but also in terms of what you learn from the two

    regions

  - should coordinate closely with Junaid and Ali on the description of

    experimental details (to avoid needless overlap) and on the complementarity

    of the forward (Junaid and Ali) and backward (Alicia) data sets

Short discussion between Nacer and Richard on theta* discrepancy

- Richard uses average_kinematics.f to make his average kinematic tables

  - need an error-weighted average over not only the different SHMS settings,

    but also over all epsilons

  - average_kinematics.f also uses eps_n_theta.f to find theta* and epsilon

    values for each (Q2bar, Wbar), the theta* values don't come from hcana

  - the cross check between theta* values in calc_xsect.f is mainly to be sure

    everything is done consistently, both values come from the same code and

    should agree

- Nacer was taking theta* values from data directly, not calculating them in

  this way

  - calc_xsect.f overwrites these theta* values, the expt values are not used

  - this method should be fine, as long as the (Q2bar, Wbar) values are

    averaged correctly

  - Nacer will modify his averaging routine (python) and do some checks

- Garth briefly mentions new CLAS-12 K+ results from Q2=0.3 to 4.5 GeV2

  arXiv: 2505.12030

  - they see strong dependence in Sigma/Lambda ratio, similar to what we can

    tell from Richard and Nacer data

    - Sigma nearly the same size as Lambda peak at low Q2

    - Sigma/Lambda ratio drops rapidly with increasing Q2

Next Meeting

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

- Thur June 5 @ 15:30 Eastern/13:30 Regina

  - KaonLT will go first

  - lots of slides to show, so the meeting might go a bit longer than usual