The estimation of peak-flow corresponding to a given return period is a significant factor in designing hydraulic structures. There are many methods used to estimate the peak-flow. In this study, two methods were used to estimate the peak-flow. Firstly, an appropriate probability distribution function is fitted to the recorded annual maximum (AM) wadi-flow series to determine wadi-flow rate with a certain exceedance probability. Secondly, in the absence of long-term wadi-flow data, peak-flows simulated by rainfall-runoff model are used. In this study, these two methods were used to estimate the differences between the design peak-flows in the Wadi AlKhoud catchment area. For the use of hydrological modelling, Intensity-Duration- Frequency (IDF) curves were developed by using General Extreme Value Probability Distribution (GEV) function. KolmogorovSmirnov test was used for testing the goodness of GEV fit with observed data. Comparison of IDF curves developed for the Wadi Al-Khoud area and the once presented in Highway Design Manual in Oman (2010) indicated that the difference between the IDF curves becomes larger as the return period increases. Geospatial Hydrologic Modeling Extension (HEC-GeoHMS) and the rainfallrunoff model (HEC-HMS) were used for delineating the catchment area and simulating rainfall-wadi flow relation. The 10-year peak-flow estimated by the observed wadiflow records is 503.37 m3 /s, which is much different from the average peak-flows of the simulated 10 scenarios (2877.82 m3 /s). This difference can be attributed to the absence of the long-term rainfall and wadi-flow data for the probability estimations and the inability to capture the spatial distribution of the rainfall over a large catchment area as Wadi Al-Khoud catchment.

Keywords : Annual Maximum Wadi-Flow, GEV Distribution, HEC-HMS, IDF Curves, Peak-Flow, Wadi Al-Khoud