Speaker
Description
Optical follow-up observations for IceCube high-energy neutrinos provide valuable insights into transient neutrino sources. However, their scientific interpretation depends critically on the analysis framework. Identifying a convincing counterpart or placing meaningful constraints from a non-detection requires a quantitative understanding of background contamination and detection efficiency. In this talk, I will present a data-driven framework for modeling backgrounds in optical transient searches, using an optical counterpart search for an IceCube multiplet event as a case study. We evaluate background contamination and detection efficiency using ZTF archival data from large background regions. This allows us to define selection thresholds in advance, without looking at the optical data in the neutrino arrival direction through blind analysis. I will also discuss how a blind analysis helps reduce bias in candidate selection and improves the reliability of background estimates.
