Le glissement de terrain et les inondations catastrophiques de la Josefina en Équateur

Abstract

The landslide and catastrophic flooding of the Josefina in Ecuador On the 29th of March 1993, a large landslide with an estimated volume of 20 million m3, took place in Ecuador, damming the Paute river. This landslide caused a number of deaths estimated between 35 and more of 100. Between the 29th of March and the 1st of May, a reservoir with a volume of near 200 million m3 formed upstream of the blockage, flooding almost 1 000 hectares of fertile and occupated land and destroying roads, railways and a regional thermoelectric plant. On the 23rd of April the water started to flow out of the dam through a channel which had been excavated in the dam in order to reduce the further damage. This in turn prompted the evacuaction of the 14,000 habitants of the valley downstream, and caused great anxiety among those responsible for the Amaluza dam, situated 60 Km downstream, which produces between the 60 and the 75% of all the electric energy consumed in Ecuador. On Saturday May 1, the river's discharge rose dramatically, nearing the 10,000 m3/s mark, and exceeding the predictions of most experts. This large flow carried large rocks and destroyed everything in its path like roads, houses, bridges, etc. only leaving standing the Amaluza dam after moments of frightening suspense. This work describes some of the geotechnical and hydraulical aspects of the Josefina landslide, the rupture of the dam and the routing of the flood wave. The various methods used to predict the rupture time and peak discharge of the flood are also compared. Three methods were employed: empirical procedures, scale modelling and numerical modelling. These predictions were of vital importance to everything situated downstream. There were strong differences between the various predictions made by the experts. In addition, some aspects of the crisis management of the reconstruction and of the natural disaster prevention are considered. The rupture of the Josefina dam is an event of first magnitude where most of the parameters were known. The experience gained must be used to improve the knowledge of this kind of events and to reduce the uncertainty of the prediction methods of large dam failure.