The UKs Contribution to the Galaxy Serpent Tabletop Exercise

This paper summarizes the United Kingdom’s (UK) contributionto the Galaxy Serpent Virtual Tabletop Exercise. The UKteam consisted of representatives from the Atomic WeaponsEstablishment (AWE), National Nuclear Laboratory (NNL), andthe University of Liverpool. Each institution in the UK team approachedthe Galaxy Serpent exercise independently with anumber of different methodologies: Radiochemistry and IsotopicCorrelations (AWE), Reactor Type Multivariate Analysis(NNL), and Laplacian Eigenmaps (University of Liverpool). Despitethe variety of methodologies employed, all three methodologiesprovided data that supported the same conclusion,namely that the seized fuel pin correlated at a suggestive positivelevel with a particular reactor. Other supporting sources ofreactor data, although available to the UK team, were not fullyexplored. As a full exercise data set was not provided to theUK, i.e., the seized fuel pin was not in the UK data set, it wasnot possible for the UK to conclude that the seized sample wasdefinitely in the UK data set. This paper details the methodologiesemployed by each institution, the UK team’s decision process,and an explanation of the misleading conclusion.