The Vajont Dam Disaster

Five decades ago, one of the deadliest dam breaches in history taught engineers a tragic lesson about constructing hydroelectric dams in geologically-unstable regions.

At 261.6 m (862 ft) in height, the thin concrete arch Vajont Dam, on the Vajont River in the Italian region of Friuli-Venezia Giulia a few dozen metres above its border with Veneto, was the tallest dam in the world at the time of its completion in 1959.  The dam had actually been conceived decades earlier under the rule of Mussolini to provide hydroelectric power to the booming cities of northern Italy, but due to World War II construction would not be authorised until 1943.  Despite opposition from local residents, especially those in the twin villages of the commune of Erto e Casso, construction would finally begin 1957 under the auspices of the local power monopoly SADE (Società Adriatica di Elettricità).

Even during the construction process, there were warning signs that this mountain valley in the southern Alps was unstable.  Despite three separate studies finding that the side of Monte Toc on what would become the south shore of the dam’s reservoir was unstable and liable to collapse into the valley should the reservoir be filled, SADE proceeded.  During road construction on the south side of Monte Toc in 1959, shifts and fractures were already being noticed on the mountainside.  Three test borings, however, failed to pinpoint the thin layer of clay contained within the limestone layer that topped the mountain (this layer was evidence of an earlier ancient landslide), and construction went ahead.  It would be this clay layer along which the surface layer would eventually slide off the face of the mountain and into the new lake.

The mountain had always previously managed to avoid a catastrophic collapse along this clay layer because before the filling of the Lago del Vajont reservoir, groundwater in the limestone was always able to seep into the clay.  Once the reservoir was filled in 1960 and the level of the new lake rose above where the clay layer was situated, the water in the limestone layer had nowhere else to seep into, thus saturating the layer.  In 1961, the first year that the dam was filled to a depth of 180 m (590 ft), a small amount of hillside creep was noticed (3.5 cm or 1.4 in per day); creep that accelerated as the level of the lake rose.  The amount of land creeping forward into the valley was massive – a 1 700 m (5 580 ft) long, 1 000 m (3 280 ft) wide chunk of Monte Toc.  Finally, in November, a 700 000 m3 (24.7 million cu ft) portion of the hillside slid into the reservoir, after which engineers dropped the level of the reservoir by 25% to try and stabilise the remaining portion of detached slope.  This slowed the creep but failed to arrest it.

At this point, the battle had already been lost because most of the south bank of the reservoir had been detached from the mountainside, and it was simply a matter of time before it all came crashing down.  However, because the creep had slowed once the reservoir had been drawn back down that first year of operation, the assumption was made that the speed of the landslide could be controlled simply by managing the level of the reservoir, and that as long as the creep remained below a certain speed there was no way the dam would become overtopped with water.  A bypass tunnel was built so that should the reservoir become blocked, water from the portion of Lago del Vajont above the slide could still be used to generate power in the dam, which was the main priority.  The second summer of operation, 1962, saw the reservoir slowly raised all the way to the 235 m (771 ft) level with the rate of creep increasing only slightly.  In 1963, the reservoir would again be raised into the 230s, and for most of the summer the creep per day was still below 1 cm per day.  SADE raised the level again in August to 245 m (804 ft), and the creep accelerated back to 3.5 cm (1.4 in) per day.  That move would serve to be too much for the mountain.  Even as the reservoir was being drawn back down to 235 m through September, the creep of the slide continued to accelerate.  By 9 October, the rate was 20 cm (7.9 in) per day.

Disastro_Vajont
640px-Vajont_monte_toc_frana

The scar left behind from the landslide on Monte Toc.  Source: Robmontagna,   http://commons.wikimedia.org/wiki/File:Vajont_monte_toc_frana.jpg.  Licensed under the Creative Commons Attribution-Share Alike 3.0 Unported licence.

At 10:39 in the evening on 9 October, the entire 270 million m3 (9.5 billion cu ft) block of mountainside collapsed into the Lago del Vajont at a speed of approximately 110 km/h (68 mph), blocking the reservoir to a depth of 400 m (1 310 ft).  This massive influx of rock displaced water in two waves on either side of the slide.  One wave carried upstream from the slide for a fair distance, taking out a hotel where 54 dam workers had been staying and killing everyone inside.  The upstream wave would also affect Erto e Casso, destroying Casso (260 m/850 ft above the lake surface) and causing some amount of damage in Erto.  The other wave headed downstream and overtopped the dam.  30 million m3 (1 billion cu ft) of water – about 26 percent of the reservoir’s water volume – would crest over the dam in a 250 m-high (820 ft) wave (the dam itself remained intact other than a meter or so of masonry work at the very top) and head down the Vajont valley into the valley of the Piave River.  In the path of this giant were the commune of Longarone and the neighbouring villages of Pirago, Rivalta, Villanova, Codissago, and Faè.  Within a handful of minutes, they would all be wiped out.  Between the seven destroyed villages, the death toll was approximately 2 000 people (80% from Longarone alone), and nearly all buildings were wiped clean from their foundations, leaving behind only mud, rock, and rubble.

The huge scar of the landslide as seen from above blocking the Vajont valley.  The Vajont Dam can be seen at extreme left.  The remnant Lago del Vajont lies to the east of the landslide.

At left, the giant scar left on the side of Monte Toc from the landslide.  At centre, the disused Vajont Dam.  At right, the San Antonio Tunnel on the SS251 highway that avoids the Vajont Dam.

A memorial to victims of the flood located inside the highway tunnel that runs past the Vajont Dam.

The Vajont Dam stands to this day, looming hundreds of metres above the residents of the valley floor.  With the collapsing hillside swallowing the valley, no attempt was made to excavate the valley or reconstruct the reservoir.   The current form of Lago del Vajont lies to the east of the landslide, dammed in by the massive wall of rock and soil left by the slide.  In order to keep the lake level constant and prevent water from accumulating behind the landslide, a pumping station transfers water from the lake down into the Piave River valley below via the original Vajont channel.

Year of legal action would ensure following the disaster, with SADE and the Italian government attempting to pin the blame on an unavoidable natural event.  An eventual trial in Rome resulted in lenient sentencing for the main parties involved with the dam.  As for the survivors of that evening who found their homes and villages wiped off the map, many of them were eventually relocated in 1971 to a newly constructed village built a 40-minute drive to the southeast.  Fittingly, the town would be called Vajontafter the valley its people left behind.  Other survivors would return to Erto e Casso to reclaim their houses (Erto’s main street is named Via 9 October in memoriam), or settle in a rebuilt Longarone, now populated mostly by people who moved there after reconstruction.  A 2001 Italian film, Vajont – La diga del disonore, was based on the story of the disaster; the director would bring the film to the valley to have a public showing for the residents using the dam as a 140-m high (460 ft) projection screen for its premiere screening.  A memorial chapel sits beside the dam.

Further Reading

Dutch, S. (2007).  Vaiont Dam, Italy.  22 June 2007.  Available at http://www4.uwsp.edu/geo/projects/geoweb/participants/dutch/vtrips/vaiont.htm.  Accessed 20 May 2012.

h2g2 (2010).  Risk in Industrial Society – Some Case Histories.  13 February 2010.  Available at http://www.h2g2.com/approved_entry/A784109.  Accessed 20 May 2012.

International Year of Planet Earth (2008). Five Cautionary Tales and Five Good News Stories.  Le Scienze Web News, 11 February 2008Available at http://www.lswn.it/en/press_releases/2008/international_year_of_planet_earth_global_launch_event_12_13_february_2008.  Accessed 20 May 2012.

La Stampa (2008).  Vajont, il muro d’acqua che ha ucciso Longarone.  9 October 2008.  Available at http://www.lastampa.it/redazione/cmsSezioni/cronache/200810articoli/37192girata.asp.  Accessed 20 May 2012.

Petley, D. (2001/2008).  The Vajont (Vaiont) Landslide.  The Landslide Blog, 11 December 2008.  Available at http://www.landslideblog.org/2008/12/vaiont-vajont-landslide-of-1963.html.  Reprinted from the original at http://web.archive.org/web/20070205070651/http://www.land-man.net/vajont/vajont.html.  Accessed 20 May 2012.

Ward, S.N. and S. Day (2011).  The 1963 Landslide and Flood at Vaiont Reservoir, Italy: A tsunami ball simulation.  Italian Journal of Geoscience 130(1): 16-26.  Available at http://es.ucsc.edu/~ward/papers/Vaiont.pdf.  Accessed 20 May 2012.

Nearby Articles

2 thoughts on “The Vajont Dam Disaster


  • I recall reading several years ago an analysis of what would happen if the Hoover Dam were destroyed. As best I remember the likely death toll would be well over 100,000. This is not counting the deaths from the nuclear explosion that would be required to destroy the dam in the first place. There would be massive and permanent agricultural losses in both the United States and Mexico. With Lake Mead drained the Las Vegas metropolitan area would be desperately short of water and unable to maintain anything close to its current population.


  • Considering the amount of water they're extracting from Lake Mead already and the continued growth of population in the region, it probably won't take the Hoover Dam being destroyed to drain the lake, just a few more decades of continued usage.

Comments are closed.