Measurement based source term reconstruction for a radioactive release

Abstract

Everyday, worldwide measurement networks for radioactivity sample the air to detect radioactivity. Sometimes stations sample radioactive materials, while the source is unknown. Because of such incidents we propose and analyze a so called source term reconstruction method, in order to find the most likely source location, release time and duration and released quantity. For this the Dutch National Institute for Public Health and the Environment (RIVM) provided the opportunity to work with an air dispersion model, which uses current weather and weather predictions originating from the Dutch national weather service (KNMI). We propose a source term reconstruction method for an instantaneous release, which is based on a combination of previous literature together with an own contribution. The performance of the method is tested for different variables as sampling locations, weather conditions and measurement and model errors. Finally a real case is used to show that despite of measurement and model errors, the proposed reconstruction method is still able to find possible source terms closely to the actual source term.