Landslides

Landslide occurrences in the Aburrá Valley
Although landslides has affected the Aburrá Valley for a long time, studies of these events only began in earliest in the 1980’s. Before 1980’s, the only study carried out was that of Shlemon (1979), who described and discussed the origin of the slope deposits distributed in the northern part of the Aburrá Valley (Bello – Copacabana area).

Shlemon (1979) considered these slopes deposits could only be explained in relation to the geomorphological evolution of the area. Shlemon distinguished three slope deposits: recent flows, intermediates flows, and older flows. He considered that these deposits could reach 200 m of thickness, and interpreted that steep slopes corresponded to older landslides, and gentler slopes corresponded to recent landslides.

Differing studies focusing on the origin of the valley were carried out during the 1980’s (Pineda 1980; Toro and Velasquez, 1984; Alvarez and Trujillo, 1985; Cuadros, 1987). In these investigations, the authors proposed a tectonic origin for the valley, based on geomorphologic evidence and the presence of large and older landslides covering the valley slopes. These authors considered the first generation of deposits were associated with the N-S trending fault systems and lineaments, parallel to the direction of the valley.

After occurrence of the Villatina disaster in 1987, studies concerning the Aburrá Valley changed from a tectonic focus toward evaluation of landslide hazard. In this way, Paz and Torres (1989) investigated the influence of precipitation on landslide occurrence in the Aburrá Valley. They concluded that landslides occurred during normal precipitation events in any month of the year. Moreover, they suggested that precipitation volumes associated with landslides do not correspond to exceptional rainfall events. Most landslides occurred under normal precipitation (Fig. 3.3).

Hormaza (1991) elaborated the record of the landslides in the Aburrá Valley in the last century and discussed the causes of landslides occurrence. According to the number of landslides examined, Hormaza concluded that the landslides in the Aburrá Valley were produced by the combination of many factors, including high precipitation, creep, lack of drainage system, and human intervention. Moreover, he emphasized that a high proportion of the landslides were preceded by symptoms associated with slow landslides.

Flórez et al. (1996, 1997) evaluated the Medellín slopes on the basis of landslide hazard. They divided Medellín city into east and west slopes. For the east slope, they reported that dunite rock surfaces had the highest susceptibility to erosive processes, accounting for natural disasters such as the Media Luna, Villatina and Santo Domingo events. They determined that the principal factors driving landslide occurrence were high precipitation, human intervention, and the natural characteristics of rocks and residual soils. For the western slopes, they considered that environmental degradation of the slopes by human intervention was the principal cause of landslide occurrence.

GSM (2002) carried out deterministic hazard mapping by GIS, based on slope angle, slope aspect, and geotechnical properties of the materials. The methodology consisted of determining slope aspect by polygons and associating average geothecnical properties with each polygon. This data was integrated into a classic stability model based on rotational slides.

Historical record of landslides in the Aburrá Valley
Several landslide inventory studies have been made in the last decade (Bustamante and Echeverry, 1984; Hormaza, 1991; Rendón and Vargas, 1998; and Flórez et al., 1996). Although continued occurrence of landslides in the valley has caused great economic losses, few no studies have evaluated the economic losses. The most relevant inventory works are those of Hormaza (1991) and Saldarriaga (2002).

Hormaza (1991) compiled information from 1880 to 1988 around Medellín, Bello, Envigado and Itagui towns, using newspaper reports, historic documents, and technical reports. Saldarriaga (2002), as part of a world program for landslide inventory (DesInventar[1]) compiled information from 1900 to 2002, including all municipalities of the Aburrá Valley. This program was supported by La Red (Social Studies Network for Disaster Prevention in Latin America), which leads the landslide inventory in South America. Saldarriaga collected 9451 reports, 44% of which represent landslide events, of those, 94 % occurred in Medellín city.