Modeling of biochemical oxygen demand in the Ergene river basin with artificial neural networks

Keywords: Ergene river, water quality, artificial neural networks In this study, artificial neural network method was used by using 10 physical, chemical and biological parameters between the years 1985-2014 obtained from General Directorate For State Hydraulic Works in 4 sampling points on Ergene River mainstream. In this study, the effects of the parameters used in the determination of BOİ value were tried to be tested and therefore BOD5 estimations were made by using artificial neural network which is operated with Microsoft Excel 2003. In the study, the basin area is examined in terms of population, industry, agriculture and climate; determined the effects of measured pollutant parameters on other parameters and also relationship between parameters. It has been seen that the neural network models are working in cases where the water quality data is more stable, especially in the rivers with high pollution load and side arm number, such as Ergene, the quality of water parameters have changed very much because of that there is no success for results.