_____________________________________________________________________________________________________ *Corresponding author: Email: tariqcivil73@yahoo.com; Original Research Article Journal of Basic and Applied Research International 14(4): 265-271, 2016 ISSN: 2395-3438 (P), ISSN: 2395-3446 (O) International Knowledge Press www.ikpress.org SIMULATION OF MONTHLY FLOW FOR THE DINDER RIVER, SUDAN TARIQ MAHGOUB MOHAMED 1* AND ETTE HARRISON ETUK 2 1 Department of Civil Engineering, Khartoum College of Technology, Sudan. 2 Department of Mathematics/Computer Science, Rivers State University of Science and Technology, Port Harcourt, Nigeria. AUTHORS’ CONTRIBUTIONS This work was carried out in collaboration between both authors. Author TMM designed the study, collected the data, did the preliminary data analysis and wrote the draft. Author EHE completed the data analysis, proofread the paper and approved the final manuscript. Received: 19 th October 2015 Accepted: 14 th November 2015 Published: 26 th November 2015 __________________________________________________________________________________ ABSTRACT The level of Dinder River in Sudan is a yearly seasonal phenomenon beginning to increase from June and reaching a peak in September and reducing thereafter. Augmented Dickey-Fuller (ADF) test certifies the series as stationary. The time plot as well as the correlogram of the series contradict this stationarity hypothesis. Even a visual inspection of the series shows that it is seasonal of period 12 months. In the sequel an attempt is made to model the monthly level using the 1972 to 2009 data using seasonal autoregressive integrated moving average (SARIMA) methods. A seasonal differencing of the series proves enough for seasonality. The correlogram of the seasonally differenced series suggests some SARIMA models. The orders investigated in the grid search are: (2,0,0)x(0,1,1) 12 , (2,0,0)x(1,1,1) 12 , (1,0,0)x(0,1,1) 12 , (2,0,0)x(0,1,0) 12 , (0,0,2)x(1,1,0) 12 and (0,0,2)x(2,1,0) 12 . The SARIMA (2,0,0)x(0,1,1) 12 model is chosen as the most adequate on the basis of minimum Akaike Information Criterion (AIC). Its residuals are shown to be white noise confirming its adequacy. It may be used to simulate the series. Keywords: Streamflow; Dinder River; Sudan; SARIMA models. 1. INTRODUCTION The Dinder River is a tributary of the Blue Nile and originates from the Ethiopian Plateau West of the Lake Tana . It flows westwards, descends into the Sudanese plain, and joins the Blue Nile River below Sennar dam. The River is about 750 km long and has a drainage basin of about 16000 km 2 , receiving annually a rainfall depth of 800-850 mm, and has an estimated runoff coefficient of 0.22 [1]. The river varies considerably in width and depth, ranging from 50-400 meters in width and 1-9 meters in depth [2]. It is a seasonal river; the flow starts in June; the flood reaches its peak in September and dries out by the end of November. The River has been measured at Hillet Idris, near its mouth, from 1907 to 1951, with a record at Gwasi from 1972. The gap in the record between 1951 and 1972 was filled by means of a statistical model [3]. Throughout the investigated period (1972- 2009), the average annual flow for the Dinder River is 2.11 km 3 . The variation of annual flows at Gwasi gauge station is high; the maximum recorded was