Coming so soon after the Boxing Day tsunami, the earthquakes in the Indian Ocean are a stark reminder of the cruelty and ferocity of nature. It can be hard to comprehend their magnitude and horror without recourse to science fiction. In proposing a global early warning system to tackle tsu- namis, earthquakes, volcanoes and even “near-Earth objects”, Professor Sir David King, the Government’s Chief Scientific Adviser, has taken on an agenda which will seem unreal to many. But it is an agenda of very real concern to the Prime Minister, who has commissioned the proposal for the July meeting of the G8 countries.

Professor King’s list of hazards is a roll call of events that we do not like to think about: those with low prob-ability but potentially devastating impact. An eruption from the Cumbre Vieja volcano in the Canary Islands, for example, could cause a tsunami which would hit Britain in about six hours and America’s eastern seaboard in about ten, supposedly wiping out Boston, New York and Miami. How do you evaluate the risk of a volcano which erupts at intervals of roughly 20 to 200 years, and has not done so since 1971? How do you compare that with the risk of an asteroid collision, which wiped out the dinosaurs 65 million years ago?

The answer must be to focus on steps that are cheap and practical. It is much easier to predict volcanic eruptions, for example, than earthquakes. Thousands of people were successfully evacuated from Kola, Indonesia, in 1986 before the volcano blew. Yet only about 150 of the world’s 3,000 active volcanoes are monitored.

A tsunami warning system is already in place for the Pacific, and is being developed for the Indian Ocean. Professor King argues that it would cost only about £100 million to extend that system to the Atlantic and to provide basic monitoring of volcanoes. If that is the case, this would be expenditure within the realms of the possible.

As Professor King acknowledges, it can be easier to get the technology right than to disseminate the information to those in danger. The most tragic aspect of the events of Boxing Day was the inability of the American scientists in Hawaii to reach the right people in time. This week, officials in Indonesia, Thailand, Sri Lanka and India were issuing evacuation warnings within minutes of the earthquake. So lessons have already been learnt.

There is nothing wrong with an independent panel of scientists advising governments on these questions. But any such grouping should resist the inclination to press for expensive research, and remember that the best way to beat nature may be with local solutions, not grand designs. The Japanese Government has upgraded its building codes to reduce the impact of earthquakes. Farmers in Cameroon plant crops to prevent soil erosion and flood damage. An increasing number of disasters are weather-related. Heightened concern about disasters should be channelled into finding the most ingenious combination of the global and the local.