Proper Rainfall for Peak Flow Estimation by Integration of L-Moment Method and a hydrologic Model

ABSTRACT: Data generation in ungagged basins is one of the most important topics in hydrology. Rainfall records usually have a longer record period than river flood records. Thus it seems logically possible to derive flood patterns from rainfall patterns using an appropriate mathematical model. Duration and temporal pattern of rainfall have a great impact on simulated peak flow which is inferred from the model. In this study, through integration of L-Moment theory and the HEC-HMS hydrologic model, the impact of duration and temporal pattern of precipitation with different return periods on peak flow was decided upon. According to the frequency analysis results, generalized logistic distribution was found proper for study area hydrometric stations. Then basin peak flow was simulated with the aid of HEC-HMS model and considering depth, duration, and temporal precipitation distribution. From comparison of simulation results with L-Moment regional frequency analysis, it was deduced that for a two-year return period, a precipitation duration between 12 to 18 hours is appropriate; and for 5, 10, 25, 50 and 100-year return periods, appropriate durations are 12, 9-12, 6-9, 3-6, and 1hour(s) respectively.