ENVIRONMETRICS Environmetrics 1999^ 00]240Ð260 Received 0 October 0887 Copyright Þ 1999 John Wiley + Sons\ Ltd[ Accepted 6 November 0888 Bayesian uncertainty estimation methodology applied to air pollution modelling Renata Romanowicz 0 \ Helen Higson 1 and Ian Teasdale 2 0 IENS\ Lancaster University\ Lancaster LA03YQ\ U[K[ 1 Cambrid`e Environmental Research Consultants\2 Kin`s Parade\ Cambrid`e CB10SJ\ U[K[ 2 Math En`ine plc\ Oxford Centre for Innovation\ Oxford OX19JX\ U[K[ SUMMARY The aim of the study is an uncertainty analysis of an air dispersion model[ The model used is described in NRPB!R80 "Clarke\ 0868#\ a model for short and medium range dispersion of radionuclides released into the atmosphere[ Uncertainties in the model predictions arise both from the uncertainty of the input variables and the model simpli_cations\ resulting in parameter uncertainty[ The uncertainty of the predictions is well described by the credibility intervals of the predictions "prediction limits#\ which in turn are derived from the distribution of the predictions[ The methodology for estimating this distribution consists of running multiple simulations of the model for discrete values of input parameters following some assumed random distributions[ The value of the prediction limits lies in their objectivity[ However\ they depend on the assumed input distributions and their ranges "as do the model results#[ Hence the choice of distributions is very important for the reliability of the uncertainty analysis[ In this work\ the choice of input distributions is analysed from the point of view of the reliability of the predictive uncertainty of the model[ An analysis of the in~uence of di}erent assumptions regarding model input parameters is performed[ Of the parameters investigated "i[e[ roughness length\ release height\ wind ~uctuation coe.cient and wind speed#\ the model showed the greatest sensitivity to wind speed values[ A major in~uence on the results of the stability condition speci_cation is also demonstrated[ Copyright Þ 1999 John Wiley + Sons\ Ltd[ KEY WORDS] Gaussian air dispersion model^ sensitivity analysis^ Bayesian uncertainty estimation^ like! lihood functions^ prior and posterior probability density functions^ prediction errors^ prediction limits 0[ INTRODUCTION This work addresses the problem of the uncertainty of predictions of an air pollution model and their dependence on the uncertainties of observations[ The uncertainty in atmospheric dispersion modelling can be a result of both the uncertainty in modelled atmospheric processes and obser! vation errors\ and the structural and numerical errors of the mathematical model[ Structural  Correspondence to] R[ Romanowicz\ IENS\ Lancaster University\ Lancaster LA0 3YQ\ U[K[ e!mail] R[RomanowiczÝlancaster[ac[uk