Workshop 6 on scattering and oscillations, November 19, 2020
Minutes by Peter

Kevin McFarland on cross section measurements and oscillations:
1. Important input is neutrino scattering data sets within the same experiment (near detector) and dedicated experiments. This always relies on a model.
2. Measure the lepton as well as possible, but it will never be enough; LAr can help with this.
3. LAr struggles with neutrons because LAr is slow
4. Converting numu from the ND to nue at the FD is an additional complication
5. Need a model and can connect it with neutrino and electron scattering data because you don't measure the whole phase space. This feedback process is not automatic.
6. Example: MINERvA: dip region problem. Tuned a small number of parameters and a number of other measurements were improved.
7. Having a capable detector with many different observables is necessary.
8. NOvA is similar to MINERvA which seems promising, but T2K seems to require a bit more work.

Q&A:
Hiro: MiniBooNE CCQE issue, what will be the next unknown unknwon? Kevin: nue vs numu differences which needs reliable theory inputs. Also models may not have all the features that we want. Ex: global fermi gas models probably need to go away, but there could be deeper problems. Hiro: is the currently envisioned program adequate? Kevin: I think the current program will struggle to make the direct measurements, and we don't have something to comprehensively validate the interaction models.

Kendall: What conversations are we missing that we need to have? There will be a number of cross section workshops, but what could we add? Information flow?

Mike Wilking: Beyond CCQE and a bit of resonance discussed, we'll also need DIS for DUNE. What do we need for the transition region? Kevin: PRISM is good providing novel measurements. Mike: is there a benefit for GArTPC? Kevin: Things always rely on models, so yes there is a role.

Jeremy: numu vs nue difference, muon storage ring addresses this problem.

Ulrich: 2p2h MINERvA and NOvA did a similar thing with this and not as much for T2K probably due to E dependence. The energy dependence seems to be related to the structure functions. High multiplicities doesn't necessarily mean DIS, it could also be FSI. Kevin: multiple energies is helpful so combining SBND up to the highest energ of the DUNE flux.

Luke Pickering on PRISM:
1. Moving the ND changes the peak from 2 GeV to 0.5 GeV
2. Can use linear algebra to reconstruct the flux shaped like a probability and then a direct comparison can be done.
3. Small corrections need where the inferred oscillated flux at ND don't match with the expected.
4. If Enutrue -> Enuobs is wrong, PRISM can largely handle this.
5. Can also measure cross sections in various regimes.
6. Can also handle nue given a numu/nue correction from a model; can address this with numu/nue at the ND. This will be nue stats limited. Related, can also fit the wrong sign component.
7. nuPRISM for T2HK moves up and down. Conceptually very similar.

Q&A:
Jeremy: nue part: why not nue far compared to nue near? Luke: stats.
Rob: Linear algebra takes flux model as an input, what if it's wrong and can PRISM help? Luke: DUNE PRISM probably can't help a lot. Mary: we expect the flux uncertainties will go down to 5% from external measurements.
Peter: the problems at 3-4 GeV are a function of Fourier series, if oscillation probabilities aren't 3 flavor will we miss things? Luke: We could consider different functions. Mary: MINOS has covered a lot of the space in numu disappearance. Mike: We can use extra on-axis data for the high energy part.

Kendall Mahn on electron-nucleus scattering
1. Electron scattering gives us the vector part of the V-A scattering. Also gives better kinematics and precision.
2. Outgoing pion spectrum don't do too well in neutrino experiments, electron scattering can help validate FSI modelling
3. Discussion is needed on what kinds of precision is needed

Q&A:
Hiro: limitations in FSI models; is electron scattering actually applicable to neutrinos? Kendall: E4Nu case with GENIE for both electrons and neutrinos.
Ulrich: In generators FSI is incoherent, but electron scattering would need some coherency. Electron scattering cuts should be selected to reflect this.
Kevin: Capabilities for neutrons?
Ulrich: An old experiment by Will Rooks with electron scattering data. Hiro: This is an inclusive data set of various particles? Ulrich: yes. The data is from 2005. One was run at HERA. One should also get in touch with the EIC community.