Article

Ranking uncertainties in atmospheric dispersion modelling following the accidental release of radioactive material

¹ Met Office, Exeter, UK
² EEAE/NCSRD – Greek Atomic Energy Commission)/National Center for Scientific Research “Demokritos”, Agia Paraskevi, Greece
³ PHE – Public Health England, Didcot, UK
⁴ IRSN – Institute for Radiation Protection and Nuclear Safety, Fontenay-aux-Roses, France
⁵ KNMI – Royal Netherlands Meteorological Institute, De Bilt, The Netherlands
⁶ BfS – Federal Office for Radiation Protection, Neuherberg, Germany
⁷ NMI – The Norwegian Meteorological Institute, Oslo, Norway
⁸ EK – Centre for Energy Research, Budapest, Hungary
⁹ PHIMECA Engineering, Clermont-Ferrand, France
¹⁰ DTU Wind Energy, Roskilde, Denmark
¹¹ RIVM – National Institute for Public Health and the Environment, Bilthoven, The Netherlands

^* Corresponding author: susan.leadbetter@metoffice.gov.uk

Abstract

During the pre-release and early phase of an accidental release of radionuclides into the atmosphere there are few or no measurements, and dispersion models are used to assess the consequences and assist in determining appropriate countermeasures. However, uncertainties are high during this early phase and it is important to characterise these uncertainties and, if possible, include them in any dispersion modelling. In this paper we examine three sources of uncertainty in dispersion modelling; uncertainty in the source term, uncertainty in the meteorological information used to drive the dispersion model and intrinsic uncertainty within the dispersion model. We also explore the possibility of ranking these uncertainties dependent on their impact on the dispersion model outputs.