Abstract. A parameter estimation procedure, sequential uncertainty domain parameter fitting (SUFI), is presented and has the following characteristics. The procedure is sequential in nature, meaning that one more iteration can always be made before choosing the final estimates. The procedure has a Bayesian framework, indicating that the method operates within uncertainty domains (prior, posterior) associated with each parameter. The procedure is a fitting procedure, conditioning the unknown parameter estimates on an array of observed values. Finally, the procedure is iterative, requiring a stopping rule which is provided by a critical value of a goal function. Performance of the SUFI parameter estimation procedure is demonstrated using three examples of increasing complexity: (1) analysis of a solute breakthrough curve measured in the laboratory during steady state water flow, (2) estimation of the unsaturated soil hydraulic parameters from a transient drainage experiment carried out in a 6-m deep lysimeter, and (3) estimation of selected flow and transport parameters from a hypothetical ring infiltrometer experiment. The procedure was found to be general, stable, and always convergent.

A data worth model is presented for the analysis of alternative sampling schemes in a special project where decisions have to be made under uncertainty. This model is part of a comprehensive risk analysis algorithm with the acronym BUDA. The statistical framework in BUDA is Bayesian in nature and incorporates both parameter uncertainty and natural variability. In BUDA a project iterates among the analyst, the decision maker, and the field work. As part of the analysis, a data worth model calculates the value of a data campaign before the actual field work, thereby allowing the identification of an optimum data collection scheme. A goal function which depicts the objectives of a project is used to discriminate among different alternatives. A Latin hypercubo sampling scheme is used to propagate parameter uncertainties to the goal function. In our example the uncertain parameters are the parameters which describe the geostatistical properties of saturated hydraulic conductivity in a Molasse environment. Our results indicated that falling to account for parameter uncertainty produces unrealistically optimistic results, while ignoring the spatial structure can lead to an inefficient use of the existing data.

Environmental data often have features that are distinct from data in other branches of science. These features include spatial and/or temporal auto‐correlation, natural heterogeneity, measurement errors, small sample sizes, and simultaneous existence of different types and qualities of data. Realistic environmental modeling requires simulation procedures that account for all of these features. In this study, a model of uncertainty analysis, BUDA, is used to account for the noted features and provide a unified framework for quantification, propagation, and reduction of uncertainty. The BUDA model is used to analyze the development of a chloride plume around an old landfill to the year 2020. This article describes the different components of BUDA as they relate to the landfill application.

No abstract