Study on synthetic and unit hydrographs by using GIS and artificial intelligence techniques

In this study, unit hydrograph and synthetic unit hydrograph parameters which are qp, tp, tb are calculated by using Snyder's, Mockus, SCS (Soil Conservation Service), and DSI (State Hydraulic Works) methods. First, calculations are done according to observed data. Then other methods mentioned above, which are based on both topographic map and geographic information systems (GIS) values, are used. Three catchments which are Damlıca, Vize, and Kumdere are studied. Snyder's, Mockus, SCS and DSI methods are applied in each catchment. The geomorphologic parameters of Damlıca catchment are determined by using geographic information systems. It is shown that, the geomorphologic parameters of the Damlıca catchment obtained using GIS are much more precise than those produced by conventional methods. Linear Genetic Programming (LGP) is also proposed in predicting daily time series of river flow data. Auto regressive (AR) technique is also presented as conventional time-series model of the same discharge data. The performance of each model was compared based on the well-known statistical performance measures. The results of each model were tabulated and illustrated in time-series diagrams. Snyder's based synthetic UHs were developed by using both digitized map and digital elevation model of a case study of a small catchment in Turkey. Multi output neural network (MONN) technique was applied to predict the three UH parameters: peak discharge (qp), time to peak (tp) and time base (tb) of a number of UHs observed in the catchment based on most relevant geomorphologic and meteorological parameters. MONN was observed to outperform the conventional synthetic UH methods. The impact of the proposed MONN is that it predicts the three parameters of the UH based on a single model compared to the conventional NN technique which utilizes a model each parameter.