DRAGON NSI

A 2021 IceCube NSI analysis (wiki, Thomas-DRAGON-2021).

Performed by Thomas Ehrhardt.

Used 3 years of DeepCore data from 5.6 to 100 GeV to search for NC NSI (Thomas-DRAGON-2021).

Performed in two parts (Thomas-DRAGON-2021):

When testing for any of the five NSI parameters, assume all others are zero (results in model dependence, required computationally).
Use a different parameterization with three NSI parameters, allowing some limits to be placed on a fully free NSI hypothesis.

The second parameterization looks like

$$ \begin{align*} H_\text{mat}(x) &= Q_\text{rel}U_\text{mat}D_\text{mat}(x)U_\text{mat}^\dagger Q_\text{rel}^\dagger \end{align*} $$

It can be approximated and reduced to a version called the "generalized matter potential" that has three parameters:

$\epsilon_\oplus$
$\varphi_{12}$
$\varphi_{13}$

Obtained frequentist confidence intervals due to "the computational infeasibility of a Feldman-Cousins approach" (Thomas-DRAGON-2021).

Succeeded by the OscNext NSI analysis.