From: paolo.desiati Subject: [icecube-c] ehe unblinding proposal Date: 2007”N4ŒŽ19“-Aú 8:21:39:JST$)B Hi Shigeru et al., I finally had some time to look at the EHE unblinding proposal and I still have some questions to ask. As I already told I am concerned about the muon bundle model you are using. The comparison between the event-summed NPE from muon bundles and from single muons seems to deviate in the saturation region (slide 18). Muons in a bundle have lower energy than the bundle itself and therefore they sample different regime in the stochastic energy loss. The energy of muon bundles in the Elbert formula depends on the mass number of primary cosmic ray. What mass are you using in the evaluation of your background. And what cosmic ray spectrum do you use ? As Henrike has clearly shown in her monopole analysis EHE/UHE searches are sensitive to high energy cosmic ray composition above 10^8 GeV where uncertainties are very high and where we do not even correctly model the primary cosmic ray spectrum. Not to mention the uncertainties on hadronic interaction models and charm production. Before unblinding you should make an evaluation of the systematic uncertainties on your background. I think this is very important since your selection cut is at zero background but you are not sure how uncertain is the zero background location. The risk is that after unblinding it will be impossible to interpret any result we get. The IC9 detector response saturates for muon energies of ~ 5*10^8 GeV. Above, energy information is lost. Maybe I missed it from your unblinding proposal but I did not see a plot or series of plot showing that we do correctly simulate saturation as the real detector does. And from this point of view the standard candle calibration does not help since we cannot simulate it correctly. What about the systematic uncertainty on the simulated signal due to the detector response modeling ? Did you include the so-called Golden DOM in your simulation ? And apparently we have a problem in the energy calibration. Another question I have is : are you thinking to publish the result ? Thanks a lot Paolo