PROBABILITY-DISTRIBUTIONS FOR FLOOD WARNING SYSTEMS

Engineering design, optimal operation, evaluation of performance, and benefit-cost analysis of a local flood warning system require an expli cit and extensive characterization of uncertainties in terms of probab ility distributions. Such a characterization is obtained via a Bayesia n Processor of Forecasts (BPF) which provides (1) a prior description of uncertainty about flood occurrence and crest height, (2) a stochast ic characterization of the forecaster in terms of likelihood functions , and (3) a posterior description of uncertainty about flood occurrenc e and crest height, conditional on a flood crest forecast. The theoret ical novelty of our BPF is that a posterior distribution can be constr ucted for any prior distribution, parametric or nonparametric, the gen erality essential in light of the variety of models used as flood cres t distributions. The conceptual novelty of the BPF opens a new researc h paradigm (which provides distributions for real-time decision making based on forecasts) that adjoins the classical flood frequency analys is (which has provided probability distributions for planning and desi gn).