International Research Journal of Public and Environmental Health Vol.1 (3),pp. 54-62, May 2014 Available online at http://www.journalissues.org/irjpeh/ © 2014 Journal Issues ISSN 2360-8803 Original Research Paper Influence of considering tributaries to reconstruct flood hydrograph of an extreme event on the upstream portion of Medjerda River Accepted 30 March. 2014 Sahar Abidi *1 , Hajji Olfa 1 , Hermassi Taoufik 2 and Habaieb Hamadi 1 1 Rural Engineering Water and Forest Department, National Agronomy Institute-Tunis (INAT), Tunisia. 2 National Institute for Research in Rural Engineering, Water and Forest (INRGREF), Tunisia. *Corresponding Author Email: sahar.abidi@yahoo.fr Tel.: +21621439755 In Tunisia, flood problems arise from the only perennial river, Medjerda, in particular the Plain of Bou Salem where two tributaries (Mellegue and Tessa) and the main station of Jendouba discharge into. The inundation of January 2003 was a catastrophic event in the Bou Salem watershed. Hence this event was chosen for research to test the reconstitution of hydrographs by different methods and to analyze the effect of its tributaries. Two models are used; Muskingum and Regression to simulate the hydrograph at different calculation times (2, 4, 6 and 8 h). To evaluate the quality of reconstitution models, three performance criteria were selected: standard deviation error ‘S’, peak relative error ‘S1’ and peak time difference ‘S2’. The reconstitution of Bou Salem’ floods (downstream station) from Jendouba (upstream station) was set in four scenarios: (1) ignoring the tributaries, (2) considering the tributary Mellegue, (3) taking into account the tributary Tessa and (4) considering both tributaries of Tessa and Mellegue. Following this application, the fourth scenario reproduced well the flood peak and illustrated how important the consideration of tributaries was. Regression model gave the best values of performance criteria and simulated the discharges hydrographs well. Key words: Flood, Mejerda River, reconstitution, tributaries, regression, Muskingum INTRODUCTION Flood risk may be defined as the probability that floods of given intensity and loss occurs in a certain area within a specified time period (Apel et al., 2006; Merz et al., 2007). Many natural disasters have led mankind to reflect on the power and frequency of floods. In the past, interest was mainly on the knowledge of maximum values of the water depth and flow. However, currently, interest is switching more and more to knowing the characteristics of hydrographs and their variations during flood propagation, particularly under the influence of tributaries. Models of flood propagation varr; there are a big number of algorithms from simple statistical receipt to the partial differential equations of Saint-Venant (Bentura, 1996). Numerical solutions of the Saint Venant equations are used to predict flood arrival time and its magnitude at various locations along a river, once the flood hydrograph at an upstream location is known (Mujumdar, 2001). To reconstitute a flood’s hydrographs, the simplest models used until now are the Muskingum and Regression models (Sahar, 2011). The Medjerda, the major and only perennial river in Tunisia, has experienced several catastrophic floods. The Plain of Jendouba-Bou Salem is at risk of flooding. It has been invaded by water at least three times; in 1973, 2000 and 2003. These floods have reminded all of the acute dangers of floods to the society and the environment. On 10th and 11th January 2003, total rainfall recorded at Jendouba, Mellegue, Tessa, and Bou Salem was 84, 103.5, 51.5 and 67 mm, respectively. Consequently the Medjrda River flooded the Plain of Jendouba – Bou Salem. The attributed runoff exceeded the maximum fill of the Mellegue Dam at 2800 m 3 /s on 11 January 2003. The river’s