Recently, Artificial Neural Network (ANN) methods, which have been successfully applied in many fields, have been considered for a large number of reliable streamflow estimation and modeling studies for the design and project planning of hydraulic structures. The present study aimed to model the rainfall-runoff relationship using different ANN methods. The Nergizlik Dam, located in the Seyhan sub-basin and one of the important basins in Turkey, was chosen as the study area. Analyses were carried out based on streamflow estimation with the help of observed precipitation and runoff data at certain time intervals. Feed Forward Backpropagation Neural Network (FFBPNN) and Generalized Regression Neural Network (GRNN) methods were adopted, and obtained results were compared with Multiple Linear Regression (MLR) method, which is accepted as the traditional method. Also, the models were performed using three different transfer functions to create optimum ANN modeling. As a result of the study, it was seen that ANN methods showed statistically good results in rainfall-runoff modeling, and the developed models can be successfully applied in the estimation of average monthly flows.

We investigated the propagation of shock waves in a prismatic rectangular channel with a horizontal wet bed. Saltwater was used as a Newtonian fluid within the entire channel instead of normal water for representing the different density fluids. It aims to point out seawater where tsunamis occur as an extreme example of shock waves. The shock waves were generated by sudden lifting of a vertical gate that separated a reservoir and a downstream channel with three different tailwater depths. The experimental data were digitized using image processing techniques. Furthermore, the flow was numerically solved by using Reynolds Averaged Navier-Stokes (RANS) equations and a DualSPHysics program (a code version of smoothed particle hydrodynamics (SPH)). After sudden removal of the vertical gate the propagations of shock waves were experimentally examined via image processing, which can yield both free surface profiles at several times and variations of flow depth with time at four specified locations. Solution successes of two different numerical methods for this rapidly varied unsteady flow are tested by comparing the laboratory data. The results indicate that the disagreements on graphs of time evolutions of water levels obtained from two numerical simulations decrease when the initial tailwater levels increase.

ÖzetHidrolojik planlamalarda eksik olan akım verilerinin tahmin edilmesi su yapılarının tasarım süreçlerinin çok önemli bir aşaması olmaktadır. Bu çalışmada Ülkemizin en önemli sel havzalarından biri olarak kabul edilen Hatay suları havzası'nda bulunan 1907 numaralı Asi nehri-Demirköprü AGİ'ye ait eksik aylık akım verileri havzadaki yakın diğer istasyonların akım verileri kullanılarak, İleri beslemeli geri yayınımlı yapay sinir ağları (İBGYSA) yöntemi yardımıyla ayrı ayrı modellenmiştir. Her bir model sonucu çoklu doğrusal regresyon (ÇDR) ve çoklu doğrusal olmayan regresyon (ÇDOR) yöntemleri ile karşılaştırılmıştır. Çalışma sonucunda İBGYSA yönteminin ÇDR ve ÇDOR yöntemlerine göre az da olsa daha iyi sonuçlar verdiği görülmüştür.Anahtar Kelimeler: Yapay sinir ağları (YSA), Çoklu regresyon, Demirköprü. Using of Artificial Neural Network (ANN) for Setting Estimation Model of Missing

No abstract

In recent years, with the effect of the climate change, drought is accepted as one of the most important natural disasters. In the planning, development and management processes of water resources, studies on the analysis of past droughts and the decreasing of possible negative effects in the future, have become even more substantial. The best adaptation to drought risk can only be achieved by adopting holistic approaches. In this study, Dörtyol-Erzin Plain, which is located in the south of Turkey and covers the fertile agricultural lands of the Asi River Basin with a drainage area of approximately 7800 km2, was preferred as the case study for hydrological drought analysis. In the literature, it is stated that there is a slow drought progress for the Asi River Basin. It is highlighted that decreasing trend in groundwater and increasing trend in evaporation and temperature parameters are remarked. Since the agricultural irrigation of Dörtyol-Erzin Plain is dependent on groundwater and surface resources, hydrological drought analysis over the long period will be beneficial for the future studies. Accordingly, Streamflow Drought Index (SDI) method was used for the hydrological drought analysis by using 35 years of flow data between the years of 1986-2020. The open source “SPI_SL_6.exe” program via National Drought Mitigation Center (NDMC) was operated in the calculations. Drought results were analyzed at different time scales of 3, 6, 12, 24 and 48 months, afterwards relevant graphs and tables were created. Consequently, the longest dry period has been determined between 2008 and 2012 water years, while the wet period has been evaluated between 2003 and 2007 ones. Furthermore, it is concluded that SDI values decreased as the monthly time periods increased, while the maximum indice values were obtained with SDI-3 in all drought periods. When all graphs are examined detailed, it can be expressed long-term droughts for certain water years are notable.