ND Physics Sample
The "ND Physics Sample" is a sample that uses a series of cuts to make it "as close as possible" to a conservative definition of the future oscillation spectrum. At its most basic, the oscillation analysis uses a best-estimate of the neutrino energy based on the lepton energy and the hadronic energy. How to calculate this varies depending on a number of factors, but the simplest is to do E_nu = E_lep + E_had. This means we need to reconstruct the muon energy and the hadronic energy.
The muon energy can either be reconstructed by range if it is contained within ND-LAr, or if its endpoint is contained within the TMS. The hadronic energy is usually estimated as the total visible energy not associated with the lepton. This can only be reconstructed if all the hadrons are contained within ND-LAr. Since this is a bit complex to check, we use an alternate definition that has been shown to be about the same thing.
LAr has 3 regions we care about
- Active region - The region that the ND-LAr can detect
- Fiducial region - dx,dy,dz is 50cm inside active region, downstream is 150cm*
- Shell region - AKA collar, which 30cm from all edges of the active region. If more than 30 MeV of hadronic E is measured in there, the hadronic component is said to be uncontained
*This is 11 radiation lengths of photons, and is to make sure we see nu+e interactions if the photon reacts further downstream. See this talk by Chris for more information.
On top of that, the hadronic containment cut is that <30 MeV total hadronic energy was deposited in the shell region. This was chosen by some study (not sure which exactly) to give us mostly contained hadrons for the purposes of calculating E_had
50cm upstream and 150 downstream cut:
-
Kevin's comment
- Chris's comment
- Kiyoung's comment 30 MeV in 30cm outermost
-
Kiyoung's talk, slide 11
- See also Chris's talk, slide 29. It was 20 MeV back then
- Muon must end:
- Before the last scinillator plane
- At least 40cm from top and bottom of the scinillator
- At least 1 bar width from the edge of the hexagon defined by the 100% U/V view overlap
- Muon should enter through front face of TMS
The TMS can only measure muons by range if they stop within the TMS. So the first three cuts relate to making sure the muon endpoint is inside the TMS. See also Andrew's talk at the the Sept 18 ND Reco/Sim meeting
If the muon was reconstucted in the last plane, then we have no information about whether the muon stopped in the last plane or left the TMS. So we instead use the last plane as a veto to reject outgoing muons
We need to do a study to make sure the reconstruction has good enough reco efficiency to use only the last plane as a veto. In some cases, an outgoing muon may be reconstructed to end in, for example, the second last plane because its last hit was missed. If that happens enough, we'd need to change the cut to second last plane or worse.
The nominal y resolution is about 30cm. So we add a cut of 40cm to reject any muons which might've left the TMS. This is to be very conservative. With muon matching, kalman filter, and the possibility of x planes, the y resolution is expected to improve.
Note that if this cut is reduced in the future, we will need to take into account the triagular profile of the scintillator planes. So at best we'll get a 5cm cut or so
The U and V views are rotated slightly. Because of this, there is only a hexagonal structure where the overlap is 100%, see figure above. It's only in this region that we'd get the ideal muon length measurement. The last bar in the region is used as a veto to reject muons which escape out the sides, like the last scinillator plane cut.
Muons coming in through the top and bottom may go through an unknown amount of material. Plus their starting direction for matching may be more poor because of the longer distance and addditional material. This was discussed as a possibly unnecessary cut because the amount of steel should be well known. But it's included for now because we're being conservative