Lb2L1520mue V2 Tuples

This involves a full reprocessing of datasets to create the V2 tuples. This replaces the previous productions v0r0p4017934 and v0r0p3866957 and they will be archived upon completion of this production.

Summary of changes

• Added all years to production
• 36 background modes added (32 for L1520me, 4 for L1520me)
• Added PHSP simulation for signal/norm mode for comparison
• Added pKJps,ee simulation/data for calibration studies (updated bu2llK_nopid to include eeLine2 NOPID version)
• me, mm, ee SS data sample added
• Removed 40% of variables through reducing # of DTF versions and removing non-useful variables
• Fixed bugs in variable naming and missing relinfo tools

Known Errors in DV from Pipeline jobs

• 2015 Trigger errors: Failed to add Hlt selection name Hlt2RecSummary to its container, well known issue due to 2015 trigger differences. Not concerning
• Signal/normalisation signal: Relations/Rec/ProtoP/Upstream relation table missing. This is a product of doing internal davinci truth matching for these modes. This is not a relaton table of concern as we use Relations/Rec/ProtoP/Charged
• InvalidMomentum, due to invalid particles in montecarlo, to be expected.

Sample Tables

Data

Decay	Stripping
\Lambda_b \rightarrow pK\mu^\pm e^\mp	Bu2LLK_meLine
\Lambda_b \rightarrow pK\mu^\pm \mu^\mp	Bu2LLK_mmLine
\Lambda_b \rightarrow pKe^\pm e^\mp	Bu2LLK_eeLine2
\Lambda_b \rightarrow pK\mu^\pm e^\pm	Bu2LLK_meSSLine
\Lambda_b \rightarrow pK\mu^\pm \mu^\pm	Bu2LLK_mmSSLine
\Lambda_b \rightarrow pKe^\pm e^\pm	Bu2LLK_eeSSLine2

Using stripping versions:

• 2011: 21r1p2
• 2012: 21r0p2
• 2015: 24r2
• 2016: 28r2 (Not latest due to no change in 2021 restripping campaign)
• 2017: 29r2p1 (^)
• 2018: 34r0p1 (^)

Signal/Norm/Calib Simulation

Decay	EventType	Stripping
\Lambda_b \rightarrow \Lambda (1520)\mu^\pm e^\mp (\Lambda (1520) forced to decay \rightarrow pK with PHSP model)	15314000	Bu2LLK_meLine NOPID
\Lambda_b \rightarrow pK\mu^\pm e^\mp (Full PHSP decay, m_{pK} spectra especially)	15314030	Bu2LLK_meLine NOPID
\Lambda_b \rightarrow pK\ J/\psi(\rightarrow \mu^\pm \mu^\mp) (\Lambda^* decays according to XLL2 model, with rough m_{pK} shape)	15144059	Bu2LLK_mmLine NOPID
\Lambda_b \rightarrow pK J/\psi(\rightarrow \mu^\pm \mu^\mp) (Full PHSP model, suitable for reweighting)	15144001	Bu2LLK_mmLine NOPID
\Lambda_b \rightarrow pK J/\psi(\rightarrow e^\pm e^\mp) (Full PHSP model, suitable for reweighting)	15154001	Bu2LLK_eeLine2 NOPID

Summary of Background Samples

(Not formatted fully, would take too long!)

pK\mu e Samples

• Lb -> pKJPs(-> MM,EE)
• Lb -> pKPsi(-> MM,EE)
• Lb -> pKMM, Lb -> pKEE
• Lb -> Lc(-> L(1520) l' nu') l nu
• Lb -> Lc(-> L(1520) pi+) l nu
• Lb -> Lc(-> L(1520) l nu) pi+
• Lb -> Lc(-> L(1520) pi+) l nu
• Lb -> Lc*(-> pD0(-> K l nu)) pi-
• Lb -> pD0(-> K l' nu') l nu
• Semileptonic modes from B2XEMU (NOTE: These come from different stripping versions)
  • B0 -> D-(-> K* l nu) l nu
  • B0 -> D*(2010)-(-> D-(-> K* l nu)pi0) l nu
  • B0 -> D*2(2460)-(-> D~0(-> K+ l nu)pi-) l nu
  • Bs -> D*s2(2573)-(-> D~0(-> K+ l nu)K-) l nu
• Lb -> pK K+ K- (PHSP)
• Lb -> pK pi+ pi- (PHSP)
• Lb -> pK K+ pi- (PHSP)
• Lb -> pK Psi (-> JPsiPiPi)
• B+ -> JPsi K+

pK\mu\mu Samples

• B+ -> JPsi K+
• B0 -> JPsi K*
• Bs -> JPsi KK
• Bd,s -> JPsi pipi

Comment on Projected Production Size ~ 5.5TB

The previous productions, v0r0p4017934 and v0r0p3866957, totalled ~ 500 GB (accounting for our variable reduction), with only 2016 and 2012, so it is to be expected that this has increased to ~ 1.2TB. With the extra SS data and addition of ee mode multiplying the amount of data samples by a factor of 3 as well as new simulation this brings the amount up to ~2.7TB.

This means the production size has been doubled by the inclusion of background samples. These are especially large due to the lack of truth matching at DV level which leads to extensive amounts of multi-candidates after nopid restripping. We don't want to implement this "online" for the background samples as we don't want to bias our background study towards where we "expect" background is breaking through our selection.