Analysis of Recorded Inflow Data of Ataturk Reservoir

Abstract

Since the beginning of the last century, the competition for water resources has intensified dramatically, especially between countries that have no agreements in place for water resources that they share. Such is the situation with the Euphrates River which flows through three countries (Turkey, Syria, and Iraq) and represents the main water resource for these countries. Therefore, the comprehensive hydrologic investigation needed to derive optimal operations requires reliable forecasts. This study aims to analysis and create a forecasting model for data generation from Turkey perspective by using the recorded inflow data of Ataturk reservoir for the period (Oct. 1961 - Sep. 2009). Based on 49 years of real inflow data from the Euphrates River recorded at Ataturk, a spilt-sample approach was adopted for testing homogeneity. The autoregressive model of order one [AR(1)] was found to be the best for the forecasting as it accurately reproduced the means, standard deviations, and skewness coefficients observed in the generated records forecast at the Ataturk reservoir. Ten sets of 100 years data have been forecasted. In Iraq, optimization of the operation of all reservoirs is necessary after operating new reservoirs in Turkey.