Presenter/Author Information

Nele Schuwirth
Christian Stamm
Peter Reichert

Keywords

multi-criteria decision support, water quality, risk attitude, river management

Start Date

1-7-2012 12:00 AM

Abstract

Decisions in environmental management can be challenging, amongst other things, due to two major sources of uncertainty. Predictions of consequences of different management alternatives can be very uncertain. Furthermore, uncertainty exists regarding the subjective preferences of decision makers and stakeholders. For a transparent decision process it is important to disentangle these different elements and the related sources of uncertainty by separating the prediction of consequences from their valuation. Predictions of consequences should be estimated as objectively as possible based on the current state of knowledge; the preference functions which are used to valuate these consequences must be elicited carefully to reflect the subjective preferences of each stakeholder. We incorporate uncertainty in decision support by applying the multi-attribute value and utility theory. We propagate uncertainty in the prediction of consequences to the valuation (numerically implemented by Monte Carlo simulation) and incorporate the risk attitude of the stakeholders to discriminate between uncertain alternatives. Furthermore, we address uncertainties inherent in the model of the subjective preference structure by sensitivity analysis to evaluate the robustness of model results against a variation in parameters of the preference model. We illustrate this procedure with a case study on the improvement of water quality in a river catchment of the Swiss Plateau. Different management options to reduce point and non-point sources will be evaluated in this study. For deriving predictions of outcomes, expert knowledge as well as mathematical models can be used. The case study is integrated in a framework for multi-criteria water management (MCWM), which allows us to use existing assessment procedures to evaluate the ecological status of aquatic ecosystems.

COinS
 
Jul 1st, 12:00 AM

Incorporation of uncertainty in decision support to improve water quality

Decisions in environmental management can be challenging, amongst other things, due to two major sources of uncertainty. Predictions of consequences of different management alternatives can be very uncertain. Furthermore, uncertainty exists regarding the subjective preferences of decision makers and stakeholders. For a transparent decision process it is important to disentangle these different elements and the related sources of uncertainty by separating the prediction of consequences from their valuation. Predictions of consequences should be estimated as objectively as possible based on the current state of knowledge; the preference functions which are used to valuate these consequences must be elicited carefully to reflect the subjective preferences of each stakeholder. We incorporate uncertainty in decision support by applying the multi-attribute value and utility theory. We propagate uncertainty in the prediction of consequences to the valuation (numerically implemented by Monte Carlo simulation) and incorporate the risk attitude of the stakeholders to discriminate between uncertain alternatives. Furthermore, we address uncertainties inherent in the model of the subjective preference structure by sensitivity analysis to evaluate the robustness of model results against a variation in parameters of the preference model. We illustrate this procedure with a case study on the improvement of water quality in a river catchment of the Swiss Plateau. Different management options to reduce point and non-point sources will be evaluated in this study. For deriving predictions of outcomes, expert knowledge as well as mathematical models can be used. The case study is integrated in a framework for multi-criteria water management (MCWM), which allows us to use existing assessment procedures to evaluate the ecological status of aquatic ecosystems.