DANGER ZONES
The whole of the Northeast, the Himalayas, Kutch and parts of the region near the Aravallis fall in high-risk zones. 59 per cent of India’s land area could face moderate to severe quakes

As another earthquake hit the country last week, this time in Delhi and adjoining areas, the need for an effective quake forecast system has once again come under focus.

Scientists have been trying for years to develop a model, which would be able to predict earthquakes well in advance. While there has been considerable success in mapping places that are prone to earthquakes, hardly anything is known about the timing and intensity of the quakes.

Scientists say there is a need to build a model for forecasting earthquakes. Key to developing any such model is the study of ‘precursors’ or the symptoms that precede an earthquake.

It has been observed that there are changes in a number of physical and electrical properties of the earth’s surface—like increase in seismic activity, radon gas emission, change in ground level or ground-water levels, electrical conductivity—before an earthquake. For example, post-quake studies have revealed that the earth’s temperature had been rising steadily in Bhuj since January 21, 2001, five days before the devastating quake in Kutch in Gujarat that killed more than 12,000 people.

Scientists believe that by working on such precursors, they would be able to develop a mechanism to predict earthquakes. In India, the National Geophysical Research Institute (NGRI) has been carrying out precursor studies for many years now and has even reported some success in predicting quakes under controlled conditions. But a comprehensive warning system is still some distance away.

NGRI director Dr V.P. Dimri, who has done extensive studies on precursors and authored a couple of books on the subject, says there are too many variables to be studied and not one can be said to be more reliable than the other.

“Changes in one or a few variables need not be conclusive signal of an impending earthquake. It’s much more complicated than that. But we can certainly build on a model by studying these variables,” Dimri said. “Moreover, even if we are able to develop a successful model, it is always going to be location specific. What works in Maharashtra might not work in the Himalayas,” he said.

Scientists at Dimri’s institute have, in fact, been successful in making accurate predictions under controlled conditions. Koyna Dam in the Western Ghats in Maharashtra is a region of continuous seismic activity and is a readymade laboratory to study earthquakes. In May 2006, NGRI, which set up nine monitoring stations near the dam, predicted a quake of magnitude 4 within the next 15 days in a radius of about 10 km from the dam. Just three days later, a quake measuring 4.2 on Richter scale struck exactly in the area predicted.

However, there has been only one instance of accurate prediction—in China way back in 1975. An earthquake of magnitude 7.3 was predicted in Haicheng a day before on the basis of changes in ground water level, foreshock activities, animal behaviour and elevation in ground level. The area was immediately evacuated. However, the failure in predicting two earthquakes in China the very next year pointed out the need of detailed research on the behavior of precursors.