Jan 22-23/26 PionLT/KaonLT Analysis Meeting Notes ------------------------------------------------- (Notes by GH) Today: Hall C meeting slides will be discussed first Thursday: Present ----------------- Regina - Garth Huber, Nermin Sadoun, Nacer Hamdi, Muhammad Junaid, Nathan Heinrich, Alicia Postuma CUA - Sameer Jain FIU - Pete Markowitz Virginia - Richard Trotta Nacer ----- Went through his draft slides for Hall C meeting next week Junaid ------ Went through his draft slides for Hall C meeting next week Sameer ------ KaonLT CoinBlocking - had some questions about EventType=4,7 requirement, which selects COIN events for the analysis - applied acceptance cuts from Richard's thesis - investigation of Q2=3.0 W=3.14 right SHMS=6.49deg setting, which has a CoinBlock correction not on the same line vs CoinRate as the Center and Left settings. Left corr=0.94-95, Main group=0.97-1.0 - when applying PID cuts, the correction goes from 0.95 to 0.85, in line with the others - Nathan notices that it's not that the low outlier points go up to join the others, as that the others have gone down from near 1 to 0.85-0.9 - since the low points are for a SHMS setting which is very close to the beamline, Nathan expects this effect is real and these cuts should not be applied - Nathan does not use PID cuts in his CoinBlock analysis. Had variations in the CoinBlock correction away from the main trend vs CoinRate, but this turned out to be real, because it make the NormYield vs Rate correlation more flat - Nathan suggests to check the S1X rates for the right SHMS setting compared to center and left. If they are substantially higher for right than the others, it's probably a real effect Friday ------ Regina - Garth Huber, Nacer Hamdi, Alicia Postuma, Muhammad Junaid, Nathan Heinrich, Vijay Kumar, Nermin Sadoun Virginia - Richard Trotta JLab - Dave Gaskell JMU - Gabriel Niculescu, Ioana Niculescu Glasgow - Kathleen Ramage Vijay ----- Went through his draft slides for Hall C meeting next week Unfortunately, we did not have a chance to review Chi Kin's slides. Richard will communicate with him on this. Richard ------- KaonLT high Q2 analysis - redid Q2=3.0 W=2.32 diamond as discussed last week - high epsilon right SHMS setting had a missing bottom right corner, the new (smaller) diamond removes this corner for the other settings too - rechecking t-binning given the smaller diamond Nathan ------ PionLT Q2=3.85 W=2.62 LT-sep - investigated CoinTiming shifts between his analysis and Junaid's - found that some timing offsets were different - also found that Junaid's RFTime cut was different - planning to use Junaid's RFTime cuts, but not planning to use his CoinTime offsets - will make more plots and recheck - Junaid will send his latest interation parameters so Nathan can see if he reproduces his sig_UNS as a bug check - t-rebinning (to gain 1 more bin at low -t) will be explored after the cross section checks are done - upper left corner of diamond will also be made slightly tighter, as was discussed in this week's PhD committee meetings Kathleen -------- PionLT LD2 analysis - still setting up iFarm replays - has a copy of Das' PID report and code - has set up meetings with Nermin, Wednesdays every other week Nermin ------ PionLT LD2 PID study Q2=1.60 W=3.08 9.17GeV - applied SHMS, HMS acceptance cuts - applied very low Cherenkov cuts for now - no cuts on SHMS calorimeter for pi+ data - Garth suggests to compare the SHMS calorimeter plots vs MM for pi+ and pi-, the pion distribution is presumably the same for both, but the electron blob will partly obscure the pi- blob - will need to use the pi+ distribution to estimate the #pi- that are removed by the electron cut - Alicia comments that with all cuts the #pions in the MM peak goes down a lot - after some discussion, it's probably due to randoms being reduced with CoinTime cut - Nathan and Garth suggested to look at random-subtracted MM w/ and w/o RFTime cut, then look to see whether you are losing more background or signal - the right side of the RFTime cut looks a bit tight, suggest to loosen to ~3.5 - Gabriel suggests to make plots also of what events fail a cut and see if there is a peak in the MM spectrum for these failed events. That will be helpful info for tuning the cut Next steps - will read more about the random subtraction procedure and also look at some pi- data Alicia ------ KaonLT u-channel shape study - setting up framework for shape study for variables other than MM - will apply the MC normalization from MM fit to other variables and compare w/data - u-comparison for Q2=3.0 W=3.14 looks good - Garth asks to recheck that recon_hcana calculates u in the same way as in hcana, and not just t - Q2, W - a few differences at edges of distribution, which hopefully will be resolved after some iterations - epsilon plots shown as well Next steps - will look at the setting where Pythia is NOT well behaved, to see if the xpfp cut discussed last week screws up the other variables Gabriel ------- KaonLT Lambda/Sigma and Lambda/Lambda* ratios - GN and IN have come up with a sound procedure for getting the ratios from the data - reminder: beam energy variations are applied to SIMC 1. Obtain Lambda and Sigma yields for MM cut around peak Y_Sigma_data, Y_Lambda_data 2. Do same for simulation, both for MM cut around peak and no MM cut and find kLambda kSigma=MM cut efficiency from MC 3. Use MC ratios method sig_Sigma/sig_Lambda = (Y_Sigma_data * (1+kSigma) *Y_Lambda_MC*Sigma_Sigma_MC) / (Y_Lambda_data * (1+kLambda)*Y_Sigma_MC *Sigma_Lambda_MC) / Next Meetings ------------------ - Thurs: Jan 29 @ 16:00 Eastern/15:00 Regina - PionLT will go first - Fri: Jan 30 @ 11:00 Eastern/10:00 Regina - we will continue where we left off - Richard will make notes for next 2 weeks, while Garth is away - will likely cancel Feb 12 meeting, as Regina group will be at conference in Banff, but can still have the meeting on Feb 13