No abstract

No abstract

This study describes a statistical approach of watercourses hydrological regimes in flood, taking into account the latter duration d and return period T. The choice of Middle Cheliff watershed as study area is linked to disasters strong return period in the western region of Algeria. The Midlle Cheliff catchment basin, located in northwest Algeria, has particularly experienced severe floods over the last years. In view of the recurrence of these unusual events, the estimation and the predetermination of floods extreme quantiles are a strategic axis for prevention against floods in this region. The a curves are first of all locally determined, directly from a statistical analysis of flow continuously exceeded during a duration d (QCXd) on different durations from available data of the study region. Then, these curves are compared to those obtained by application of different regional models VFS (Vandenesse, Florac and Soyans) in which two indices of the watershed characteristic flood are taken into account, a descriptive duration of the flood dynamics (D) and the instantaneous maximal annual flow of 10 year return period (QIXA10). The final choice of the model is based on verification of certain criteria, such as: Nash and the root mean squared error (RMSE). The closest regional models to the local ones are Florac’s for low duration and return periods, and Vandenesse’s for large return periods, for different durations. These results could be used to build regional Q-d-F curves on ungauged or partially gauged Algerian basins.

In the management of water resources in different hydro-systems it is important to evaluate and predict the sediment load in rivers. It is difficult to obtain an effective and fast estimation of sediment load by artificial neural network without avoiding over-fitting of the training data. The present paper comprises the comparison of a multi-layer perception network once with non-regularized network and the other with regularized network using the Early Stopping technique to estimate and forecast suspended sediment load in the Isser River, upstream of Beni Amran reservoir, northern Algeria. The study was carried out on daily sediment discharge and water discharge data of 30 years (1971–2001). The results of the Back Propagation based models were evaluated in terms of the coefficient of determination (R2) and the root mean square error (RMSE). Results of the comparison indicate that the regularizing ANN using the Early Stopping technique to avoid over-fitting performs better than non-regularized networks, and show that the overtraining in the back propagation occurs because of the complexity of the data introduced to the network.

No abstract