Flood frequency analysis in dams by downscaled data

Design flood discharge of spillway of dams is one of the important criteria for dam safety analysis and one of the main principles that the designer engineer depends on during design of dams. There is a great variety of methods used around the world to estimate of design flood discharge. In this study, design flood discharge for spillway of Surqawshan Dam located in North Iraq is estimated by using statistical downscaling and Soil conservation service (SCS) techniques. This new design flood discharge is compared with the conventional design flood discharge that was used in the design of the dam. 26 downscaled meteorological variables as input and the local monthly maximum rainfall (Rmax) data of a gauge station nearby Surqawshan Dam as output were modeled by Linear Genetic Programming (LGP) to obtain predictive equations. The flood discharge of Surqawshan Dam corresponding to 10,000 year return period (Q10,000) was determined by SCS method. The results show that Surqawshan Dam is under safety since the design flood discharge, Q10,000, was obtained to be 405 m³/s although it was computed as 1091 m³/s by using the observed peak discharges. The fundamental purpose behind this outcome is that, according to the downscaled rainfall data, there is a decreasing trend in annual maximum rainfall data.