Monday 11th May 2026
https://indico.cern.ch/event/1674975/
New LHC pO data together with Pierre Auger will “reduce model freedom at a level never reached for cosmic ray analysis”
Pierre Auger results point to high energy CRs having heavier composition than just protons, not simply an issue of hadronic modelling
QGSJET excluded by Pierre Auger results? (slide 14)
Confirm shift in X_max related to elasticity, i.e. diffractive component? Would be nice to constrain elasticity as a function of CoM energy. Need more data on diffraction in pA/AA systems. Forward measurements important to constrain Xmax in proton-oxygen
EPOS LHC → EPOS LHC.R takes into account new data and new knowledge from EPOS 4. Global approach taking into account collective effects. EPOS LHC had a bug in nuclear fragment production, fixed in LHC.R
ATLAS measurement confirms EPOS LHC.R cross-section. Statement from Tanguy that a more inclusive measurement would be desirable, for example down to pT > 0.1 GeV.
Statement from Tanguy that it would be very interesting to measure nuclear fragmentation
One of the main sources of uncertainty now is pion-air interaction
No data for pion-Air. Problems with pion cross-section/inelasticity
Neutron tag in ZDC → gives 50% purity of pion-Air
Heavily impacted by turning hadronic rescattering on/off
Q&A:
H Menjo: What causes the reduction in cross section in new EPOS?
T Pierog: I tuned to new data. Also related to the introduction of colour transparency. Overall reduced cross section.
D. d’Enterria: I’m concerned about the cross section for pO. Why didn’t ATLAS lower the measurement to 100 MeV. My impression is that there is a bias from the choice of 500 MeV and larger extrapolation. We did not see this in CMS when requiring forward/backward activity which is more inclusive.
TP: The introduction of colour transparency should have more impact on this type of analysis.
DdE: The differences between different models of the oxygen profile is +/-10–15 mb. I think we should try to look at much more inclusive observables. I’m surprised by the spread of predictions.
TP: I think these differences come from diffraction.
DdE: I agree that we don’t know how diffraction works for pA.
Angantyr has improved Glauber calculation by including fluctuations → diffractive excitation. Doesn’t yet include collective effects.
Angantyr calculates AA interactions by stacking NN interactions on top of each other. Predicted cross-section higher than ATLAS measurement. To decrease xsection, need to increase number of NN interactions (as each individual NN xsection stays the same)
Nuclear geometry: by default Angantyr uses the general GLISSANDO parametrisation for all nuclei → explicitly not valid for A <= 16!
ATLAS result uses Pythia 8.308. New Pythia 8.317 motivated by CR cascades can use AngantyrCascade tune → gives 428 mb instead of 442 mb (compared to ATLAS’ 396 +/- 11 mb)
For oxygen, can use a harmonic oscillator model (currently untested) → gives 347 mb
Hadronic final states are key to understand small nuclei
Q&A:
J Liu: Do you have a prescription for defining systematic uncertainties on the model?
LL: The idea was that there would be some variables to change going from pp to pA but then not much to do going from pA to AA. Ask me offline and I will provide suggestions. I was surprised to see that hadronic rescattering could have such a large effect (from EPOS).
Not included in pO runs
Neutrinos that FASER measures are a probe of forward hadron production
FASER can probe enhanced strangeness model (see result on slide 12)
FASER results are already disfavouring DPMJET (at least the charm production)
Q&A:
T Sykora: Result on slide 12 - what message should I take from this? All generator predictions sit in one corner, could there be a systematic shift coming into the data?
MQM: Our data does not agree well with the generators. This is a very simplistic template fit but we do take systematic correlations into account. We let the pion and kaon fraction to float. What this plot tells me is that all generators predict a certain ratio of pions to kaons that does not agree with our data.
LHCf only installed on outgoing proton beam side
Q&A:
M Tasevsky: Are you planning to do a K0 measurement in your pp dataset?
HM: Yes this is being worked on.