Skip to main content navigation
Site logo

Predicting Landslides - Transcript

[c]
00.00            Japanese landslide Video
                      Exts Durham University Earth Sciences dept
                      Researcher at computer with Japan landslide images

Guide Voice: In August 2004, the precise moment when a landslide reached catastrophic failure was caught on camera by a team from the Japanese ministry of Land, Infrastructure & Transport. In this instance there was sufficient warning, and nobody was injured, but on average landslides kill over 8,000 people every year, mostly in the developing world.

Researchers from Durham University's International Landslide Centre visited the Japanese landslide site in order to map and measure its movement and analyse soil samples in the laboratory. At a time when we seem increasingly powerless before the ravages of nature, The International Landslide Centre has made great strides in understanding and predicting landslides.

00:44 SOT: David Petley, Reader, Dept of Geography, Durham University - "The primary aim of the Landside centre is to help developing countries in particular to deal with landslide problems that they face, and there are three main activities that we undertake. First and perhaps most important is we undertake fundamental research into landslide problems in those countries. So we go out and we measure landslides, we use lab techniques to investigate them, we try to develop schemes for assessing where landslides might happen in the future. The second activity that we do is a disaster response, so we have a team on 24 hour standby to fly out to developing countries to help them to deal with landslide problems that have occurred at very short notice. And the third activity that we undertake is to try to build up capability within developing countries to deal with landslide problems."

01.25            Laboratory shots
                      Soil samples
                      Soil sample into Back pressured shear box
                      Results on screen two shot researchers

Guide Voice: In order to understand the nature of landslides they have developed a number of research techniques of their own. The analysis of soil samples taken from within landslides has taught them a great deal about how a landslide is set off. They designed and built their own "Back pressured shear box" to simulate conditions within a landslide. Rainfall is the major factor in the majority of landslides that are not caused by earthquakes. So inside this device the water pressure is gradually increased within the soil sample, and its behaviour is then monitored on a computer.

001.57 SOT David Petley (over cu graph on screen pointed out) - "So this is the water pressure increasing simulating the rainfall and as the rain falls onto the soil and sinks through the water pressure goes up. This is the movement of the landslide initially there's no movement and round about here we start to get a little bit of movement, the landslide starting to slide starts to accelerate and then suddenly around about this point it suddenly fails and the landslide accelerates down the slope to a catastrophic failure."

02.22            Group Laser scanning in field
                      CU scanner operated
                      Scanner turning
                      3D models on computer

Guide Voice: Another technique they have developed is to use lasers to scan a suspected landslide site. The laser tracks back and forth across a slope to build up an exact image. This can then be mapped into a detailed 3D model. Any subsequent movement is then detected by comparing a new scan to the original model. By interpreting the pattern of movement in a slope at an early stage, they can then extrapolate its likely behaviour in the future:

02.49 SOT: David Petley - "It's quite exciting because we are now in a position to predict pretty much exactly when a landslide might happen and how fast it might be when it does occur. This is done really in a very simple way, we measure the movement of the landslide, perhaps just by putting pegs into the slope and measuring the movement of the pegs and we can interpret the movement pattern to tell us what the landslide might do in the future and under certain conditions we can actually predict exactly when the slope might move and whether it will be a very catastrophic or a very rapid failure or a slow movement."

03.21            Landslide photos on computer screen
                      Researcher in vision
                      CU photos buried roads and lorry stuck

Guide Voice: In the case of small scale landslides a few hours warning could save lives, and the Landslide Centre believe that in the case of large scale slides they could predict them days or even weeks ahead. This life saving expertise is currently being deployed in a number of countries in the developing world,and is also being used to mitigate the impact of landslides by advising on the construction and sighting of roads in areas like Nepal, where 70% of roads are destroyed within 3 years.

ends (Dur: 3 mins.46 secs.)
Page contact: Tom Abbott Last revised: Fri 1 Apr 2005
Back to top of page