This space intentionally left blank.
John Kelley, UW-Madison, January 2007
Initial comparisons (see this page) between unweighted trigger-level PTD and Photonics-simulated nusim Monte Carlo for 2005 indicated fairly good agreement. There was some indication of a different energy response between the two models (more low-energy events with Photonics), but the overall number of events agreed to a few percent.
However, this "small" energy response difference is exacerbated when reweighting the original E^{-1} spectrum to an atmospheric neutrino spectrum, resulting in a rather large normalization difference. We present here some comparisons at trigger level and at level 4 filtering, under the following conditions:
Executive summary: there is a normalization difference of 1.5 between PTD and Photonics for atmospheric neutrino MC, and this persists from trigger level, to level 4, and finally to tight point-source cuts (not shown here, but observed by Jim Braun). There is also a noticeable difference in the shape of the zenith angle spectrum near the horizon (more events in Photonics). These differences are significantly reduced if we scale the absolute OM sensitivites when using Photonics by 75%, as originally observed by Markus Ackermann.
|
|
|
|
|
|
|
|
This space intentionally left blank. |
For information on 2005 filtering, please see David Boersma's page. Level 4 is L3 + the recoos paraboloid reconstruction.
 |
 |
 |
 |
 |
 |
 |
 |
As it turns out, Markus Ackermann made similar observations when performing PTD / Photonics comparisons using data and MC from previous years. He found that one has to scale the canonical values of the OM sensitivities by 75% to achieve a reasonable agreement between data, Photonics, and PTD. We have reproduced this result for MC using the detailed wl06v200 Photonics tables and released aluminum-fix Amasim (as described in the introduction):
 |
 |
Figure 1: True neutrino zenith angle comparison of atmospheric-weighted
nusim
for 2005 using Photonics, PTD, and Photonics + absolute OM sensitivities scaled
by 75% (ratio to PTD on right).
The 75% value seems to be a bit too low, as one can also see with a plot of the energy spectra:
 |
 |
Figure 2: True neutrino energy of atmospheric-weighted nusim
for 2005 using Photonics, PTD, and Photonics + 75% OM sensitivities (ratio to
PTD on right).
But, clearly, the combination of canonical OM sensitivities with the released Photonics tables is problematic for atmospheric neutrino simulation. We plan to refine these comparisons and are also studying the impact of relative OM sensitivities, as Stephan Hundertmark and others have demonstrated good agreement with low-level data in 2003 [physto.se] by varying the latter.
Please feel free to contact me if you have any questions or comments.