RESUMO
In the very early stages of response to an accidental release of radioactivity leading to environmental contamination, it is likely that only limited measurements of radioactivity in the local environment will be available on which to base decisions concerning protection measures and radiation monitoring activities. Model predictions will be used to aid understanding of the radiological situation and to form a basis for emergency health protection decisions. This paper presents an analysis of the relative importance of contributors to the imprecision associated with emergency response calculations based on a few off-site measurements, using predictions from the UK Met Office's NAME III (Numerical Atmospheric dispersion Modelling Environment version 5.2) dispersion model. The results presented extend those from a previous study in which a simple Gaussian plume model was used and confirm the key parameters contributing to imprecision. The potential extent of the sheltering countermeasure resulting from a hypothetical release in real weather conditions occurring in 2007 and 2008 is also presented.