━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
                NOTES FROM ALI_MINI MEETING (2023-06-25)

                         Richard L. Trotta III
               ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━





◊ Look into absolute error on target thickness


◊ HGCer efficiency

  ◊ Peter said I didn't need to use HGCer for kaon cut


  ◊ SIMC HGCer cut, hgcer_eff*weight as the correction


  ◊ SIMC aerogel tray cut


  ◊ HGCer for two regions, in each region plot avg photo-electrons as a function of the momentum

    ◊ Poisson distrobution was used to get the actual efficiency for given number of photo-electrons


  ◊ For each region of the HGCer, Avg number of photo-electrons for all runs of a momentum setting. Then plot this avg vs momentum in each region.

    ◊ Apply hole cut


    ◊ Plot NPE of each region


    ◊ Root can provide the avg NPE.


    ◊ Repeat for each run of a momentum setting and get the total avg for that momentum setting


    ◊ Modify HGCer eff script to get NPE of each region per run. Make a CSV file with momentum, avg NPE, run number. Use CSV to get avg NPE per momentum and fit with a Poisson to get the efficiency (by integrating under the curve??).


◊ Luminosity

  ◊ BCM correction factor: if current < 60 then bcmcorr = 1.0+0.045 * (log(60.) - log(current)) /(log(60.) - log(2.))


  ◊ BCM correction factor: if current > 60 then bcmcorr = 1.00+0.010 * (current - 60.) / 25.


  ◊ Apply to current as current = current * bcmcorr


  ◊ Plot EL_CLEAN with no cuts, should be very stable


  ◊ Peter had scalers (EL_CLEAN) within 0.1-0.2%


  ◊ Plot production with HMS EL_CLEAN vs current for settings with a variety of currents. Study each individually (even dummy can be used) and the boiling should be obvious.


◊ Kaon PID

  ◊ For pi/proton peaks in MM_k, simulate in SIMC and then subtract


  ◊ Aerogel

    ◊ make sure geometric cuts match in data and SIMC and don't include any low efficiency regions


    ◊ use 3 NPE cut rather than 1.5 to have less protons


  ◊ HGCer

    ◊ In data, require <1 NPE


    ◊ In SIMC, remove events that have pion ID (i.e. kaon has decayed) at the aerogel position AND are outside the hgcer hole


    ◊ The simc ntuple variables that give the particle mass squared and position in spectrometer can be used to remove these kaon decay events


    ◊ This is an important correction, especially at low momentum. Quite a lot of kaons decay in the hut, but the decay pion makes a track and passes all the way to the aerogel and calorimeter.


    ◊ To be even more accurate, the "weight" can be multiplied by (1-pi_eff) IF the particle is a pion and L<21 meters or so (where 21 is the aerogel L).


    ◊ To simulate the shape vs MM of pions in SIMC, again multiply the weights by (1-pi_eff)


    ◊ In simc, you should not only run ep->(e, n, pi+), but also the channel ep->(e, pi, Delta0). The threshold for this reaction is 1.08 GeV, and so it falls under the 1115 Lambda peak. The shape is a curve that rises very quickly above 1.08 GeV, so exclusive pions would underestimate the pion background.