It is for the first time in seismic history that military personnel inside bunkers have been killed.

There have been cases before of soldiers killed in earthquake while overground. But the present case is unique and military engineers will now have to revise their design to make bunkers safe against seismic forces.

But the most frequently asked question is whether it is possible to predict the occurrence of an earthquake.

The answer to this question is, 'yes, to a sufficient degree of accuracy, about a few hours before the occurrence of an earthquake.'

Complete coverage: Tremors across borders

The earthquake precursors discussed here are exclusively for moderate to large earthquakes, meaning earthquakes of magnitude 6.0 or more on the Richter Scale. Earthquakes of lesser magnitude may or may not necessarily exhibit all the precursors.

Earthquake precursors have been divided into three broad categories. a. long term precursors. This means more than five years in advance; b. medium term precursors. This means within about four to six months in advance; and c. short term precursors. This refers to signs within 20 to 30 hours.

Long term seismic precursors: The location and magnitude of the likely epicentre is calculated based primarily on statistical studies, and the prediction is probabilistic. It is reliable when the probability of occurrence in the next five years or less is more than 0.6. However, such predictions are more useful to academic professionals and researchers.

Medium term precursors: Such predictions are able to give the location of the quake to a sufficient degree of accuracy, and the likely time of occurrence and magnitude are somewhat grotesque. The period of six months or so could be used for creating seismic awareness, taking suitable mitigatory measures, imparting seismic training to various forces etc. These are indicative of changes in the geomagnetic field, changes in the gravity field, rise in sub-surface temperature, rise in ground radioactivity etc. All these measurements indicate that something is wrong inside the earth. But it is not possible to arrive at any definite inference. Inputs from such parameters are useful to administrators and academicians.

Short term precursors: These could be seen by the common man and are highly reliable. People in seismically active areas may keep a copy of this article for further use. There are several types of short term precursors which could be seen and observed by different strata of society.

Thermal precursor: A few months before the occurrence of an earthquake the average temperature of the area keeps increasing. Weather report bulletins refer to temperatures above or below average by so many degrees. It is seen that in case the area is heading for an earthquake, the average temperature goes on increasing. On the day of the earthquake it is about 5 to 9 degrees Celsius above the average normal temperature for that day. This could be monitored by the meteorology department and also by thermometers inside homes.

Water precursor: There is a sudden rise or fall in water level in the wells. It could be as high as one metre. Sometimes the well water may turn muddy. At times a fountain appears inside the well. All these changes happen about one or three days before the earthquake. Sometimes a fountain appears in the ground. This normally happens a few hours before the quake.

The Gujarat earthquake of 2001

Another form of water precursor is seen in the flow of riverwater. There is a sudden and rapid increase or decrease of water flow in the river or nallah. This happens about one to two days before the quake.

Seismo-electromagnetic precursor: This is a very reliable precursor. It occurs and is exhibited about 10 to 20 hours before the quake. Before the occurrence of an earthquake the subsurface temperature rises. As a result of this the geomagnetic field is reduced. The reduction in geomagnetic field adversely affects the propagation of electromagnetic waves. This is experienced abundantly on the radio, television and telephone.

If one is tuning the radio at 1000 kHz, then the same station will be received in the potential epicentral area, about 10 to 30 hours before the quake, at higher frequencies. Maybe 1100, 1200, 1300, 1800, 1900, 2000 kHz or so.

Similarly, reception on television is affected by audio, visual and spectral disturbances. The telephone's reception is also adversely affected and there is a continuous disturbance that you can hear.

Apart from this, a recent reliable precursor is the mobile telephone. It has been found that about 100 to 150 minutes before the occurrence of an earthquake, mobile telephones start non-functioning or malfunctioning. The timespan indicated is sufficient to take all necessary mitigatory measures.

A word of caution: all the above precursors are valid only when seen and manifested extensively. Failure of one or two instruments should not be taken as a seismic precursor.

Special: 'Musharraf, please leave'

Animal precursor: It is seen that 10 to 20 hours before the occurrence of an earthquake, the entire animal kingdom becomes highly disturbed and restless. They move in a directionless manner and in fear. Birds do not sit on trees but move about at a low height, emitting a shrill noise. Rodents like rats, mongooses etc are in a panic. Domestic animals like cows, dogs, cats etc struggle against being tied up, and even turn on the owner.

Human precursor: Doctors and nurses observe that some sensitive patients in hospitals become highly disturbed. They exhibit a sudden rise in blood pressure, heart trouble, headache, migraine, respiratory disorders etc. Further, these psychosomatic signs are manifested without any provocation. It is also seen that the number of patients in the out-patient department increases by five to seven times, some 10 to 20 hours before the quake.

The best indicator is the number of child deliveries in any hospital. On the penultimate day of the earthquake the number of deliveries goes up about three to five times, while on the day of the earthquake it is as high as seven to eight times the normal.

Arun Bapat is a research seismologist. He was formerly Head, Earthquake Engineering Research Division at the Central Water and Power Research Station, Pune. He can be contacted at: arun_bapat@vsnl.com