Author(s): Hamish Harvey; Jim Hall
Linked Author(s):
Keywords: No Keywords
Abstract: The need for adaptive strategies to respond to uncertain long term processes of change in flooding systems is now well recognised. In the face of uncertainty, it is desirable to keep options open and enhance resilience. However, adaptive options are seldom achievable at zero marginal cost–in other words genuine‘win-win’solutions are rare. Therefore, appraisal frameworks are required that can evaluate adaptive strategies over extended timescales in the context of a wide range of alternative futures. In principle this may appear to be a relatively straightforward task. However, in practice, appraising flood risk management options, whose nature and cost will vary depending on which of many uncertain futures actually materialises, can be a computationally burdensome and conceptually taxing task. This paper describes a formal approach for specifying risk-based decision appraisal in the context of future uncertainties and long term processes of change. At the core of the analysis is a deterministic flooding model for estimating the severity and consequence of flooding, and cost model for computing the cost of flood risk management options. Around this kernel are layered risk calculations, simulations of long term change, comprehensive sensitivity analysis, decision analysis and analysis of the robustness of risk-based decisions to the ambient uncertainties. In the example presented here, processes of long term change include increasing relative mean sea level and increasing damage potential resulting from economic growth. In both cases the rates of increase are uncertain, as are the costs of implementing the interventions from which options are constructed. These various uncertainties are expressed as probability density functions (PDFs) over the variables in question. They are propagated through to option performance by means of a Monte Carlo experiment, allowing a PDF on performance to be constructed for each option.
Year: 2010