ABSTRACT
Hydrologic variables such as rainfall and runoff are highly localized in their characteristics. This chapter provides the definition of return period and a general formulation for normal, log-normal, Gumbel, exponential, Pearson, and log-Pearson distributions. It presents examples to illustrate database structure, calculations of sample statistics, determination of data skewness, predictions, and confidence intervals. Flood frequency analysis is a statistical technique applied to the long-term record collected at a gage station under a steady hydrologic condition in a watershed. There are two types of databases developed from the complete data series (CDS) for the hydrologic frequency analysis: annual maximum series (AMS) and annual exceedance series (AES). Plotting a data set on a probabilistic scale requires the determinations of relative plotting positions among the data points. Because the selection of the best-fitted probabilistic model is a trial-and-error process, the major effort is how to determine the plotting positions to display the data points on the selected probabilistic graphic paper.
