Antares/KM3NeT group meeting - reconstruction

Europe/Amsterdam
Zoom

Zoom

Reconstruction meeting format (Bouke)

We discussed and decided upon a format for the future reconstruction meetings. A summary of the most important points can be found below:

  • Frequency
    • once every two weeks, starting from mid-August
  • Accessibility
    • allow all Nikhef-KM3NeTters to join, if interested
    • Bouke will email the rest of the group to spread the word
  • Contents:
    • A usual round to discuss the on-going reconstruction analyses
    • Brían: Threshold for discussion will be kept very low; everyone is welcome to bring up analysis-details and/or points of confusion
    • Start every second reconstruction meeting with a more dedicated in-depth discussion about a reconstruction-related topic (e.g. a 30 min presentation, similar to the junior meetings)
  • Procedure:
    • Chair will write minutes with the discussed points at the end of every meeting
    • Jordan: Reconstruction meetings should be seen as an opportunity for improving the documentation and updating the wiki

Thijs - shower elongation and Jpp PDFs

Thijs showed a toy model which can be used to study the accuracy of reconstruction based on the Jpp arrival time PDFs. The toy model samples eight light emission points along the longitudinal direction of a shower, generates the resulting hit patterns, does a maximum likelihood estimation of the vertex based on the Jpp PDFs and computes the offset of the reconstructed vertex w.r.t. the MC-truth vertex in x, y and z. Whilst cm-precision is achieved in the x- and y-directions, the error in the longitudinal z-direction is typically on the order of 1 m.

  • Aart:
    • What happens to the error if the number of sampling points is increased from 8 to e.g. 100?
    • The toy model currently assumes no emission at the point of the true neutrino vertex

Jordan - PDF slider plots

Several slider plots were showed to demonstrate the influence of the elongation sampling size on the shape of the shower-PDFs in Jpp. Jordan showed how decreasing the sampling size increases the relative contribution of early hits.

  • Aart:
    • We may have to consider incorporating the effect of elongation in the shower-PDFs directly in the future
There are minutes attached to this event. Show them.