Flood Damage Probability Evaluation

Abstract

The mean recurrence interval between two exceedances is perhaps the most widely used flood risk criterion, yet it is well known that the probability distribution of total damage, and expected total damage are by far more significant criteria. In this paper a mathematical model, based on Borgman's recursive equations, is developed to compute total damage probability distribution and expected value, if the probability distributions of flood peaks, and of the damage caused by a single flood exceedance are known. The assumption is made that peak flow is the only relevant flood characteristic to flood damage. At the expense of a considerable amount of computation, solution can be also envisaged for cases where damage severity depends on the waiting time of exceedances, on the return period between successive exceedances or on season time. Annual maxima or partial duration series methods can be selected in flood probability computation. In the latter case the use of a Polya model rather than a Poisson model for exceedances occurrence is suggested.