The research plan for this Presidential Young Investigator is divided in three parts: 1) quantification of the risk of extreme rainstorms; 2) quantification of the risk of extreme floods; and 3) incorporation of the developed methodology into a risk-based decision-making framework. Based on work done on storm regionalization and a probabilistic approach to storm transposition, the first part of the investigation includes the assessment of extreme storm data base and development of estimation methods to account for record incompleteness; develop stochastic spatial models; study the feasibility of analytic asymptotic approximations to the tails of the distributions using extreme value theory; and determine the reliability of alternative estimation procedures and quantify the uncertainty of the exceedance probability estimates. The second part comprises research on the spatial variability of several basin characteristics relevant to rainfall/runoff transformations under extreme conditions; study and assess the effects of land use changes on resulting floods for the homogeneous conversion of storm data bases to floods; investigate the performance of several rainfall/runoff models for converting extreme rainfall into flood hydrographs; and integrate all findings into a method for estimating probabilities of extreme floods and assess the reliability of those estimates. The third part is to apply the developed models to real systems and to incorporate them into a risk-based decision making framework.
