ammari, abdelhadi ecole nationale supérieure d’hydraulique - gee research laboratory, Algeria , moramarco, tommaso scientific research council(cnr) - research institute for geo-hydrological protection (irpi), Italy , meddi, mohamed ecole nationale supérieure d’hydraulique - gee research laboratory, Algeria

A simple entropy-based method for discharges measurements in gauged and ungauged river sites: the case study of coastal Algerian rivers

32485

In river monitoring, it is quite difficult to sample velocity points during high floods, especially in the lower portion of flow area, due to the danger that operators might face during the measurement. In this context an important contribution is provided by the entropy theory which identified a linear relationship between the mean flow velocity and the maximum flow velocity which can be easily sampled also during high floods for its position in the upper portion of the flow area. The entropic relationship is robust and based on the estimation of a sole parameter (M). Therefore, if one was able to estimate the dependence of (M) on hydraulic river characteristics and/or morphological basin ones then it would be possible to assess the mean flow velocity and, hence, the discharge just by sampling the maximum velocity through, for instance, non-contact radar sensors, for ungauged basins or weakly gauged ones. Based on the above insights, the purpose of this work is twofold. The first, determine simple formulations between the entropy parameter (M) and morphological characteristics of the basin, subtended from the river site of interest, in terms of drainage density, drainage area and perimeter…etc. The second, identify a formulation between hydraulic river characteristics (Manning’s roughness, hydraulic slope,... etc) and the entropy parameter. Algerian gauged river sites are used as case study. That relationship to find M parameters for any sites within the basin was identified with a good correlation of 0.7 and a Mean Square Error (MSE) of 0.046. A quick validation was made as well on some other hydrometric stations in order to check the accuracy of the obtained model. A good representativity of the model was found with errors not exceeding 5% except for ALTAIRAC station (errors of 10%) which, however, need additional analysis. Five other Italian sites were used to check the model validity.

35

Entropy , discharge monitoring , ungauged sites , coastal Algiers watershed

bulletin de l’institut scientifique, section sciences de la terre

44