EXPECTED PROBABILITY AND ANNUAL DAMAGE ESTIMATORS

Controversy continues over the relative merits of traditional frequenc y estimators and the ''expected probability'' estimator of flood risk that incorporates an adjustment for parameter uncertainty. Both have s olid theoretical motivation, but address different concerns. The descr iption of hydrologic risk and uncertainty provided by new risk and unc ertainty procedures adopted by the U.S. Army Corps of Engineers, and r isk-based design procedures developed by others, are shown to be equiv alent to the expected probability model in simple cases. A 1995 Nation al Research Council (NRC) report recommended against use of the expect ed probability model for evaluating expected annual damages and the pr obability of flooding; in particular, the NRC analysis and the 1989 Ar nell analysis demonstrated that expected probability estimators yield risk and damage estimators that generally have large positive biases. Historical arguments and related issues are reviewed. Resolution of th is controversy and success of the new U.S. Army Corps of Engineers (US AGE) risk and uncertainty procedures require a clear framework for und erstanding what is meant by risk, variability, and uncertainty. Such r isk analyses can better represent a community's vulnerability to flood ing and the large uncertainty in estimates of expected damages and res idual flood risk.