Climate scientists in race to predict where natural disaster will strike next

Conference in Boulder will step up world's efforts to establish an early warning system for extreme weather events

A woman grieves near her relative's body after dozens of people were killed in landslides in China's Gansu province last week. Photograph: Aly Song/Reuters

The world's leading climate scientists will gather this week in the United States to hammer out plans to set up an early warning system that would predict future meteorological disasters caused by global warming.

The meeting, in Boulder, Colorado, has been arranged at diplomatic level amid fears that storms, hurricanes, droughts, flooding and other extreme weather events now threaten to trigger widespread devastation in coming decades. A series of meteorological catastrophes have dominated headlines in recent weeks, while scientists have warned that figures so far for this year suggest 2010 will be the hottest on record.

Recent events include a record-breaking heatwave that has seen Moscow blanketed with smog from burning peatlands, the splintering of a giant island of ice from the Greenland ice cap, and floods in Pakistan that have claimed the lives of at least 1,600 people and left 20 million homeless.

Scientists say events like these will become more severe and more frequent over the rest of the century as rising greenhouse gas emissions trap the sun's heat in the lower atmosphere and bring change to Earth's climate and weather systems. However, their ability to pinpoint exactly where and when the worst devastation will occur is still limited. The aim of the Colorado meeting is to develop more precise predictive techniques to help pinpoint the location and severity of droughts, floods, and heatwaves before they happen and so save thousands of lives.

"The events in Moscow and Pakistan are going to focus our minds very carefully when we meet in Colorado," said Peter Stott, head of climate monitoring at the UK Met Office. "On both sides of the Atlantic we have been monitoring what has been going on with the aim of understanding their precise causes so that we can provide better warnings of future disasters."

The meeting in Boulder will be the first full session of Ace, the Attribution of Climate-related Events, which has been set up by scientists from the world's three leading meteorological organisations: the US National Centre for Atmospheric Research (NCAR), the UK Met Office and the US National Oceanic and Atmospheric Administration (NOAA).

The aim, said Stott, would be to develop a modelling package that would allow scientists to forecast the kind of events that the world has been witnessing over the past few weeks – before they struck. The fact that the Foreign Office has been closely involved in setting up Ace reveals how seriously the issue is taken by politicians.

Meteorologists have developed remarkably effective techniques for predicting global climate changes caused by greenhouse gases. One paper, by Stott and Myles Allen of Oxford University, predicted in 1999, using temperature data from 1946 to 1996, that by 2010 global temperatures would rise by 0.8C from their second world war level. This is precisely what has happened.

But although meteorologists have developed powerful techniques for forecasting general climatic trends – which indicate that weather patterns will be warmer and wetter in many areas – their ability to predict specific outcomes remains limited. It is this problem that will be tackled, as a matter of urgency, at the Ace meeting in Boulder.

An example of the complexity that faces meteorologists is provided by the weather system that scorched Moscow, said Stott. "Moscow has a stable high pressure system over it, much like the one that brought a heatwave to Europe in 2003. However, for a while the land around the city acted as a natural air conditioner, keeping the air cool through evaporation of moisture from the ground. But the land eventually dried out and there was no more cooling. Hence the soaring temperatures."

To forecast an event like that, scientists need to be able to quantify all the variables involved and also develop a very precise model of the land surface, added Stott.

"These are the sorts of things we need to understand. We need to be able to forecast events weeks or months ahead of their occurrence so people can mitigate their worst impacts. We also need to consider the longer-term context and see if we need to build better sea defences at a particular location and assess how high dykes or walls need to be. Certainly, one thing is clear: there is no time to waste. The effects of global warming are already upon us."