UNCERTAINTY ANALYSIS AND VALIDATION OF ENVIRONMENTAL-MODELS - THE EMPIRICALLY BASED UNCERTAINTY ANALYSIS

In the present paper, the principles of Empirically Based Uncertainty Analysis (EBUA) are described, EBUA is based on the evaluation of 'per formance indices' that express the level of agreement between the mode l and sets of empirical independent data collected in different experi mental circumstances. Some of these indices may be used to evaluate th e confidence limits of the model output. The method is based on the st atistical analysis of the distribution of the index values and on the quantitative relationship of these values with the ratio 'experimental data/model output'. Some performance indices are described in the pre sent paper, Among these, the so called 'functional distance' (d) betwe en the logarithm of model output and the logarithm of the experimental data, defined as d(2) = Sigma(1)(n)(1n M(i) - ln O-i)(2)/n where M(i) is the ith experimental value, O-i the corresponding model evaluation and n the number of the couplets 'experimental value, predicted value ', is an important tool for the EBUA method. From the statistical dist ribution of this performance index, it is possible to infer the charac teristics of the distribution of the ratio 'experimental data/model ou tput' and, consequently to evaluate the confidence limits for the mode l predictions, This method was applied to calculate the uncertainty le vel of a model developed to predict the migration of radiocaesium in l acustrine systems, Unfortunately performance indices are affected by t he uncertainty of the experimental data used in validation, Indeed, me asurement results of environmental levels of contamination are general ly associated with large uncertainty due to the measurement and;sampli ng techniques and to the large variability in space and time of the me asured quantities. It is demonstrated that this non-desired effect, in some circumstances, may be corrected by means of simple formulae.