Development of a radionuclide data processing tool with tailored flagging criteria and analyses for
Introduction
The preparatory commission of the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) has set-up a worldwide network, called the International Monitoring System (IMS), of high precision sensors designed to detect possible violations of the Treaty. The network uses four technologies: seismic, infrasound, hydroacoustic and radionuclide identification. CTBT Member States can be assisted by a National Data Centre (NDC) to analyse and interpret the data collected by the IMS. A NDC is composed of scientific organizations with expertise in the CTBT verification technologies allowing to advise its government on the verification of the CTBT.
In Belgium, SCK CEN is one of the scientific organizations within the Belgian NDC (NDC.be). SCK CEN focusses on the analysis and interpretation of radionuclide observations in the IMS network coupled with Atmospheric Transport Modelling. For this purpose, the observations of the IMS are retrieved, processed and analysed in the context of the CTBT. Due to the strong background of some CTBT-relevant radionuclides in the atmosphere (e.g. Xe-133, Cs-137, I-131, Na-24), flagging criteria are required, certainly for small NDCs such as the NDC.be, to focus on observations with a high likelihood to be relevant for the CTBT verification. For such flagged observations, specific analyses (e.g. isotopic ratio plots, spectra summing) can be performed to further determine their CTBT-relevance.
Objectives
The objective of this Master Thesis is to develop a processing tool to: i) retrieve relevant radionuclide data from the CTBTO database (SQL), ii) flag observations based on defined criteria, iii) gather additional information for the flagged observations (e.g. State-Of-Health, detection history for comparison) and iv) generate specific analyses (e.g. isotopic ratio plots) and specific data (e.g. summed spectra) for further analysis in existing tools. The flagging criteria will be based on existing criteria (currently: percentile based) but also considering the interface with an external flagging script (e.g. machine learning script). The generation of specific analyses and specific data will be based on state-of-the-art techniques for CTBT verification and will be tested on recent CTBT-relevant observations in the IMS. The tool should be based on open-source software (e.g. Python).