Keywords: air quality, spatial assessment, modelling, uncertainty mapping, data assimilation, Bayesian, probability, exceedance, air pollution
Towards uncertainty mapping in air-quality modelling and assessment
The aim of this paper is to promote the use of uncertainty mapping when spatial assessments of air quality are made. A large number of air quality maps are produced for scientific and policy purposes but rarely are corresponding maps of their uncertainty included. The need for such maps and the methods to produce them are described. Several uncertainty parameters are discussed but it is recommended to use the probability density function as the basis of the uncertainty estimates. Several examples are provided discussing indicative uncertainty, ensemble methods, comparisons with observations, spatial representativeness, uncertainty in exceedances and probability of exceedance.