Uncertainty of mass discharge estimates from contaminated sites using a fully Bayesian framework

Mass discharge estimates are increasingly being used in the management of contaminated sites and uncertainties related to such estimates are therefore of great practical importance. We present a rigorous approach for quantifying the uncertainty in the mass discharge across a multilevel control plane. The method accounts for: (1) conceptual model uncertainty through Bayesian model averaging, (2) heterogeneity through Bayesian geostatistics with an uncertain geostatistical model, and (3) measurement uncertainty. An ensemble of unconditional steady-state plume realizations is generated through Monte Carlo simulation. By use of the Kalman Ensemble Generator, these realizations are conditioned on site-specific data. Hereby a posterior ensemble of realizations, all honouring the measured data at the control plane, is generated for each of the conceptual models considered. The ensembles from different conceptual models are combined via Bayesian model averaging, yielding the overall mass discharge probability distribution. Method features are illustrated on a synthetic test case.