Semi-analytical transient analysis of river embankments with a synthetic design hydrograph.

The design of a river embankment is usually carried on in a steady framework with reference to the river level of assigned return period. The seepage analysis results in a safety evaluation of the water pressure which stresses the embankment increasing the risk of failure both for static instability and piping effects. Nevertheless the seepage steady state condition is rarely verified in most of river embankments due to the limited duration of floods and the poor conductivity of soils usually adopted in the building of river banks. The calculation in such conditions can lead to a noneconomic design with very large and expensive river levees. Attempts to design in unsteady regime, by means of mathematical models, were carried on in recent years but the selection of the design hydrograph in the river is still an open problem. In the last year a new procedure for the identification of an efficient synthetic design hydrograph (SDH) with assigned return period was developed by Maione et al. (2000); it has been successfully applied in the design of flood control reservoirs and in the evaluation of the flood routing in rivers with large flood plain storage (Maione et al., 2001). The procedure presented by Maione et al. (2000) leads to the outline of a SDH, with given return period, starting from a database of real floods observed in a river station. The main parameters of the design floods can be determined by means of regionalization procedures so that it can be possible, for a certain river, their outline even without a database of real floods. In the present paper the a.m. new SDH is applied in the seepage analysis of a river embankment, as an attempt to design the river embankment in unsteady regime; the SDHs of given return period are applied in a semi-analytical model to a simplified river bank so that the main hydraulic quantities (maximum water level, time of maximum etc.) of the seepage in the soil can be determined. Then all the real floods are inputted in the model and for each of them the main hydraulic quantities are determined. The reliability of the design flood methodology is tested with a comparison between the outcomes from the SDHs and real floods application