Researchers at a North-East university have made a breakthrough which could save thousands of lives every year by predicting when natural disasters will occur.

The Durham University team has developed a means of precisely calculating the moment when a hillside will reach the point of catastrophic failure and produce a landslide.

On average, 8,000 people around the world are killed by landslides every year and, with the catastrophic events of the Boxing Day tsunami focusing world attention on predicting such natural disasters, it is hoped the development could lead to an early warning system which would save lives.

Dr David Petley, reader in geography, said: "We may not be able to prevent landslides, but with advance knowledge we hope we can help to reduce the human casualties."

The research team from the International Landslide Centre, formed in 2003 and based in the university's Geography Department, have been working on means of predicting landslides and are on 24-hour standby to fly out to a disaster anywhere in the world.

Landslides are most common in Third World countries, often triggered by heavy rainfall or earthquakes.

By measuring slopes and soil movements using a laser scanner and studying it on computer models in the lab using a specially-designed black box to simulate conditions within the landslide, the Durham team have developed a calculation which can predict the future behaviour of the landslide.

Dr Petley added: "Our work has now reached a very exciting stage because we are now able to predict pretty much exactly when a landslide will happen and how fast it might develop.

"This can be done in quite a simple way by putting pegs into the slope and measuring any movement.

"Under certain conditions we can predict exactly when a slope might move, and whether not it will be a rapid and catastrophic failure and or slow movement."

Members of the team are currently working around the world, including Taiwan, Nepal and Bhutan, working alongside Nasa scientists to detect landslide movements using satellite tracking.