Evaluate the Performance of stochastic models in Generating Monthly Discharges for the Rivers in North of Iraq

Abstract

The data of monthly discharge which are available for most of the rivers in the north of Iraq have a short recorded period ,which has a great influence in the design and planning of water resources systems , where this short recorded not contain a critical and extreme flows sample of the low flow sequences. Two stochastic models , bivariate "Thomas & Fiering" model and multi-side "matalas" model were used for generating time series for(100) years for the lower Zab and Diyala river at upstream of Dokan Dam and Derbendekhan Dam using the available recorded data for the period (1962-1997) . From application of different statistical tests (Kolmogrov–Smirnov), Chi-Square (X2), Root Mean Square Error (RMSE), Mean Absolute Error (MAE, Z), found that "Matalas" model is more performance than the bivariate model, "Thomas & Fiering" generate time series for both rivers, so it can be concluded that the "Matalas" model can be proposed to be used for generating the flow series for the different rivers in the northern Iraq. Key Words: Time Serious for monthly discharge, Stochastic Models , bivariate "Thomas & Fiering" model, "Matalas" model, Statistical Tests