Uncertainty analysis in design rainfall estimation due to limited data length : a case study in Qatar

This chapter presents a statistical modeling framework to quantify uncertainty in design rainfall estimation due to sampling error arising from limited data length. We used rainfall data from three stations in Qatar and adopted Monte Carlo simulation technique to carry out uncertainty analysis where bootstrapping is used to define standard error in the sample estimates of mean, standard deviation, and skewness of the observed annual maximum (AM) for the 24-h duration rainfall data. From the results of three goodness-of-fit tests it has been found that Log-Pearson Type 3 (LP3) is the most favorable distribution for the three selected stations. Results from bootstrapping show that the estimate of the mean is associated with the smallest degree of standard error, while skewness has the highest error level. By applying the developed statistical modeling framework, the confidence intervals for design rainfall are derived for 2- to 100-year return periods at the three selected stations in Qatar.