Development of Depth-Duration-Frequency Relationships Using Homogeneous Region Concept

This study presents an attempt to develop depth-duration-frequency curves for a region including South Carolina, North Carolina, and Georgia by identifying homogeneous regions. Existing methodologies of identifying the homogenous regions are reviewed and utilized in this study. The aim of the study is to update the existing intensity-duration-frequency curves and develop isopluvial maps that can be used to develop these curves at ungauged sites throughout the region. For this study, 17 durations ranging from 15 minutes to 120 hours are analyzed. For each duration, return periods of 2, 10, 25, 50, and 100 years are considered. Data is collected from various sources and the missing daily data is estimated using the surrounding daily stations. Maximum annual precipitation series is computed for each duration at each site. The scale correction factor is applied to correct the maximum precipitation data for clock hour interval recording. The serial and cross-correlation of maximum annual precipitation series are computed to verify the assumption that rainfall is independent of time and space. Outliers in the maximum annual precipitation series are identified and removed. The impact of missing periods on outliers is also investigated. The maximum annual precipitation data at each site is also fitted to candidate probability distributions in order to determine rainfall amount for each return period and duration. The distributions used for this study are normal, lognormal, generalized extreme value, Pearson type III, and log Pearson type III. The goodness-of-fit of each distribution is evaluated with the chi-squared goodness-of-fit test to choose the best fit. Isopluvial maps are developed by performing spatial analysis. The intensity-duration-frequency curves are developed and compared to the existing curves.