Limitations of preliminary flasher study

fiedl / hole-ice-study

This project aims to incorporate the effects of hole ice into the clsim photon propagation simulation of the icecube neutrino observatory.

Other

4 stars 2 forks source link

Limitations of preliminary flasher study #94

Open fiedl opened 6 years ago

fiedl commented 6 years ago

The preliminary flasher study #59 shows that a flasher scan is possible. The results, e.g. the best parameters, however, are not to be trusted, yet, as several systematics are not handled correctly in this study:

Long-scale ice properties: An outdated ice properties configuration is used.
Tilt is not simulated.
Anisotropy is not simulated.
An outdated detector geometry is used, i.e. distances between DOMs could be wrong.

fiedl commented 6 years ago

The best parameters (#84) are different from the best parameters of Martin's flasher study (#87).
Martin's best parameters (#87) are in agreement with the reference curve (#6), mine (#84) are not.
When running a flasher simulation with my simulation but Martin's best parameters, simulation and data are in disagreement (#90).
Even the best parameters with fixed r = 0.5 r_DOM (Swedish camera) have a bad LLH (#92).

fiedl commented 6 years ago

How large could the effect of the DOM displacement be on the fitted hole-ice parameter?

Data source: ~/hole-ice-study/results/flasher_parameter_scan_full_brightness_with_thinning

Compare the agreement of string 64: Good agreement for esca = 0.6m, r = 2.25 r_dom. To reduce the photons seen on string 64 to an amount seen on string 54, one needs to go to esca = 0.4m.