Keywords: complex site, Gaussian plume modelling, krypton-85, model validation, nuclear fuel reprocessing, radioactivity
Comparison between field measurements of 85Kr around the BNFL Sellafield reprocessing plant and the predictions of the NRPB R-91 and UK-ADMS atmospheric dispersion models
Air concentrations of the radioisotope krypton-85 (85Kr), released from the BNFL Sellafield reprocessing site, were compared with the predictions of two atmospheric dispersion models NRPB R-91 and the next generation UK-ADMS model (version 3). Data on stack emissions and local meteorology were used as model input, whilst building effects were included using a simple "effective stack height" approach. Both models tended to make conservative estimates of air concentrations, with slightly higher overestimations being predicted by the UK-ADMS model. Both models had similar fractions of predictions within a factor of two (35% for UK-ADMS and 34% for R-91) and five (56% for UK-ADMS and 62% for R-91) of the measurements. It was concluded that, when using the simple "effective stack height" approach, there was little discernible advantage of applying the UKADMS model in preference to the R-91 model at the BNFL Sellafield site. Further research will investigate the influence of the UK-ADMS "complex effects" modules on the dispersion predictions.