Uncertainty quantification in aerosol dynamics

The influence of uncertainty in coagulation and depositions mechanisms, as well as in the initial conditions, on the solution of the aerosol dynamic equation have been assessed using polynomial chaos theory. In this way, large uncertainties can be incorporated into the equations and their propagation as a function of space and time studied. We base our calculations on the simplified point model dynamic equation which includes coagulation and deposition removal mechanisms. Results are given for the stochastic mean aerosol density as a function of time as well as its variance. The stochastic mean and deterministic mean are shown to differ and the associated uncertainty, in the form of a sensitivity coefficient, is obtained as a function of time. In addition, we obtain the probability density function of the aerosol density and show how this varies with time. In view of the generally uncertain nature of an accidental aerosol release in a nuclear reactor accident, the polynomial chaos method is a particularly useful technique as it allows one to deal with a very large spread of input data and examine the effect this has on the quantities of interest. Convergence matters are studied and numerical values given.