A screening model to quantify resilience

Concurrent advances in statistics and biology have introduced similar concepts to these disciplines: the use of robust estimators and the investigation of resilient ecosystems, respectively. This paper and those following in this series explore the application of these concepts of water resource systems. The emphasis here is on the concept of a resilient water resource system, or one which does not respond precipitously to a major ‘surprise’ during the course of its economic life. The general flatness of the response surface in the vicinity of the global optimum is explored, leading to suggestions for selection of small‐, intermediate‐, and large‐scale system design. The conclusions indicate that systems characterized by a small number of design options are somewhat more brittle and are more sensitive to optimization procedures, while systems which contain significant redundancies are relatively insensitive to the identification of global optima and lend themselves more readily to negotiations in the institutional and political spheres.