Earthquake Prediction in China - Monitoring Animal Behavior
George Pararas Carayannis



Introduction


In order to reduce the risk of an earthquake and reduce and mitigate its effects, it is necessary to predict where and when a future, large earthquake may occur. For example, it would be important to know when such an earthquake will hit, where it will strike, and what the level of its destructiveness may be. Earthquake prediction at the present time is not an exact science, and forecasts of earthquake occurrences have not been very accurate. Presently predictions are given in statistical terms. For example, when a prediction is made that :here is a 90% chance that an earthquake will occur in the next 50 years, it does not mean that this earthquake cannot happen tomorrow or it may not be delayed by 50 years. Thus, present predictions are not within a reasonable time frame that can be of usefulness to planners, policy makers, and those in government that deal with public safety.


To understand earthquake prediction, three different time frames have been assigned by scientists: long term, intermediate and short-term predictions. Long term prediction involves a time frame of a decade or more and can only be general and with very limited usefulness for public safety. Intermediate term prediction would fall into a time span of a few weeks to a few years, and again it would not be of great practical usefulness. It is the short-term prediction, that is specific information on the time and location of an earthquake given within days, weeks, months - not years - that would be useful for any kind of public safety and evacuation.


Several specific geophysical, geological, and chemical methods are presently used for earthquake prediction. To the list of geological and geophysical events and precursors we should add one more method that has been used with much success in China: that is the monitoring the behavior of animals before quakes.



Validity of Earthquake Prediction


It was in the early evening hours of February 4, 1975, when an earthquake of magnitude of 7.3 struck the densely populated town of Heicheng, in the Liaoning Province, in northeast China. As expected, because the magnitude was great and the area highly populated, damage to Haicheng and to neighboring towns was extensive. However, what was different about this earthquake is that very few lives were lost. The reason for the low death toll was that the occurrence of this earthquake had been successfully predicted.

China\'s Seismic Zones


As early as 1970, the State Seismological Bureau in China had identified Liaoning Province as a high earthquake-prone area. A short term earthquake prediction was given to the population in the Haicheng area as early as mid-January 1975. This prediction was based on a number of geophysical observations of precursor events as well as abnormal animal behavior. When the earthquake struck, the people of Haicheng and neighboring towns had been warned, and the warning saved many lives and averted a disaster of major proportions. Four other disastrous earthquakes were predicted by Chinese scientists during the 1975/76 period, giving hope that earthquake prediction was finally possible.



When a great earthquake with magnitude of 7.8 struck the city of Tangshan on July 28, 1976, at least 655,000 people died and 780,000 more were injured. There was no prediction for this earthquake, and therefore no warning. Hope in the accurate predictability of earthquakes evaporated. Although precursor events had been observed and geophysical and geochemical anomalies had been detected, these precursor events occurred over a very widely-spread area making it extremely difficult for scientists to focus on any particular region and thus issue a short-term prediction, or a warning. There were remarkable differences between precursor events of this particular earthquake and those of other predicted earthquakes. Furthermore, no significant foreshocks were observed. The examples given illustrate the validity of earthquake prediction at the present time. However, progress is being made which may lead to better predictions.


Knowing that an earthquake will occur in any particular region is not sufficient. There is no doubt that earthquakes can occur anywhere along any of the numerous faults of China. To provide predictions too far into the future would be totally impractical. Thus, only the short-term predictions should be relied upon, since longer term predictions could have greater social and economic effects, particularly if they are false. For predictions to be of usefulness they would have to specify time, place, and magnitude of a forthcoming earthquake with sufficient precision and level of confidence to be practically useful. Studies of earthquake recurrence frequency. although useful from a statistical point of view, cannot be used with sufficient confidence for a prediction. For a prediction to be valid, it has to be sufficiently precise, and supported with a great deal of research and instrumentation. However, it is difficult, if not nearly impossible, to instrument all the faults and to monitor all the different parameters used for prediction.
Therefore, until such time as earthquake prediction becomes a more exact science, it might be best not to issue any predictions. Hopefully, it is only a question of time before effective methods can and will be used for reliable predictions. Until then, and until the complex interactions of the behavior of earthquake faults is understood, it would be best not to utilize predictions for public evacuation. But this does not mean that preventive measures cannot be taken to ensure the safety of the public and the protection of property. With proper planning and public awareness, the effects of the earthquake hazard can be mitigated.

Earthquake Prediction Research


Presently what is called prediction is not really that. It is simply scientific research on understanding the workings of earthquakes. There is not sufficient historical data on which to base the number of hypotheses that have been proposed for earthquake predictions and, therefore, there is no way to judge the ultimate success or failure of such predictions. It would be difficult to explain to the public the difference between scientific research studies and actual predictions. There will be difficulty in understanding and it would be even more confusing.


For example, the present research efforts are directed toward the prediction of the numerous small earthquakes (magnitude 3 and 4) that occur frequently. By predicting the smaller earthquakes and by understanding the earthquake process that is taking place, it is believed that the long-term goal of predicting large, destructive earthquakes will be achieved. Thus researchers install dense nets of seismograph stations along active sections of active faults. Such networks are used to establish detailed studies of the pattern of seismic activity in respect to both the location of the earthquakes and the time of their occurrence. By so doing, researchers attempt to identify gaps or anomalies, in the pattern of seismic activity. Numerous types of instruments and methods are used for earthquake prediction. These instruments measure tilts int the ground\'s surface, as well as changes in the magnetic field near active faults . Usually these changes precede the occurrence of some earthquakes.


In addition to the field investigations, laboratory and theoretical studies are also being carried out by governmental organizations and various universities in China and throughout the world. These investigations include studies of the mechanism of earthquake faulting of various rock types under conditions of high temperature and pressure, and detailed investigations of ground swelling, otherwise known as dilatancy, which precede an earthquake. Computer simulations of the conditions of the earth are being made as another means of obtaining patterns of earthquake processes in the hope of learning more about properties of the earth that control earthquakes.

The Use of Animals in Earthquake Prediction


Research being carried out in China has indicated that recognition of unusual animal behavior in a systematic way can lead and be used, in conjunction with other methods, as a means of predicting large and potentially destructive earthquakes. The following are examples of observed unusual animal behavior before major earthquakes occurred.



Unusual Animal Behavior - In 1920, the largest earthquake to hit China with a magnitude of 8.5 occurred in Haiyuan County, Ninghxia Province. According to reports of eyewitnesses, prior to this earthquake, wolves were seen running around in packs, dogs were barking unusually, and sparrows were flying around wildly. It is reported that prior to the 6.8 magnitude earthquake in 1966 in Hsingtai County, Hopei Province, in Northern China, all the dogs at a village near the epicenter had deserted their kennels and thus survived the disaster.


Prior to the earthquake of July 18, 1969, (magnitude 7.4) in the Pohai Sea, unusual behavior was observed in seagulls, sharks, and five different species of fish. Based on observations of unusual behavior of giant pandas, deer, yaks, loaches, tigers and other animals, a warning was issued at the Tientsin People\'s Park Zoo, two hours before the earthquake struck.


The Chinese began to study systematically the unusual animal behavior, and the Haicheng earthquake of February 1975 was predicted successfully as early as in mid-December of 1974. The most unusual circumstance of animal behavior was that of snakes that came out of hibernation and froze on the surface of the earth. Also a group of rats appeared. These events were succeeded by a swarm of earthquakes at the end of December 1974. During the following month, in January 1975, thousands of reports of unusual animal behavior were received from the general area. Local people saw hibernating snakes coming out from their holes and into the snow. In the first three days in February the activity intensified even more and unusual behavior of the larger animals such as cows, horses, dogs and pigs was reported. On February 4, 1975, an earthquake of magnitude 7.3 struck the Haicheng County, Liaoning Province.


More instances of unusual animal behavior were reported. A stock breeder in northern China, feeding his animals before dawn on July 28, 1976, in the area of the Kaokechuang People\'s Commune, approximately 40 kilometers away from the city of Tangshan, reported that his horses and mules instead of eating were jumping and kicking until they finally broke loose and ran outside. A few seconds later, a dazzling white flash illuminated the sky. Tremendous rumbling noises were heard as a 7.8 magnitude earthquake struck the Tangshan area.


Other reports of unusual animal behavior prior to the occurrence of earthquakes have been reported in the literature and in books. Such unusual animal behavior included goats refusing to go into pens; cats and dogs picking up their offspring and carrying them outdoors; pigs squealing strangely; chickens dashing out of the coops in the middle of the night; fish dashing about aimlessly; and birds leaving their nests. It has also been reported that zoo animals refused to go back into their shelters at night; snakes, lizards and other small mammals evacuated their underground nests; insects congregated in huge swarms near the seashores; cattle sought higher ground; domestic animals became agitated; and wild birds left their usual habitats.


Surveys done in China show that the largest number of cases of unusual animal behavior precede the earthquake, particularly in the 24 hours before it strikes. In other parts of China where major earthquakes have been preceded by foreshocks, unusual behavior in rats, fish, and snakes were observed as early as three days prior to the earthquake, but continuing to several hours, or even a few minutes before.

Studies of Animal Behavior

Throughout China\'s long history, unusual behavior has been observed in every kind of common animal. Most of the behavior falls into the category of unusual restlessness and disorientation.


Since animals have the capability of acting as predictors of earthquakes, the Chinese scientists have carried out surveys of animal behavior variations prior to earthquakes. A team of scientists including biologists, geophysicists, chemists, meteorologists, and biophysicists conducted a survey in the Tangshan area and in 400 communes in 48 counties around it after the 1976 earthquake. The scientists visited a number of places that were hit by other destructive earthquakes and, through interviews and discussions with local people, collected information on over 2,000 cases of unusual animal behavior occurring prior to an earthquake. The majority of the reports involved domestic animals. Based on this survey a preliminary report was prepared by the Chinese identifying 58 kinds of domestic and wild animals that had demonstrated unusual behavior.


The principal focus of research work in China has been on the behavior of pigeons. Biological studies on pigeons determined that a hundred tiny units exist between the tibia and fibula on a pigeon\'s leg. These nerve units are connected to the nerve center, and are very sensitive to vibrations. Scientists determined that prior to an earthquake of magnitude 4.0, which occurred in the area of the study, fifty pigeons that had severed connections between the tibia, fibula, and the nerve centers, remained calm before the earthquake, while those with normal connections became startled and flew away.


Because of the success in monitoring unusual animal behavior for the prediction of certain earthquakes, the Chinese, who have pioneered this work, have looked into ways to construct instruments that would duplicate the sensory organs of animals which were able to monitor, and sense, stimuli preceding an earthquake. Researchers found it very difficult to understand the mechanism of response stimuli. Physical or chemical stimuli come out of the earth prior to an earthquake and these must be the stimuli that animals can sense. For example, dogs may be able to hear the microfacturing of rocks a few milliseconds before a quake shock reaches the surface. Electromagnetic changes in the earth prior to an earthquake may be sensed by such animals as sharks and catfish which have low or high frequency receptors and sense such changes actively or passively. Also such electromagnetic field changes could be affecting migrating birds and the navigational ability of fish.

Mechanisms of Animal Responses

What is the sensory mechanism of animals that controls their responses to changes related to an impending earthquake? As mentioned earlier, the behavior of an animal might be subject to changes in the magnetic field preceding a major earthquake and such changes may be sensed by energy transfer at the electron level which, in turn, cause changes in the cellular behavior, or response. The living cell is essentially an electrical device and a micromolecular structure, and the sensory organs are all interconnected. Electromechanic changes occurring prior to the occurrence of a large earthquake may be sensed by certain animals and filtered, then instinctively interpreted. Thus animals may have the means and sensitivity to sort out and discriminate the threatening precursory signals of an impending earthquake, thus activating a behavior pattern for survival.


These precursory electromagnetic or electromechanic changes which precede an earthquake, although mixed with background noise, must be filtered by animals and coordinated through their sensory response to the total environment. Thus, behavior is determined by the sensitivity of the different component parts of the living system to the surrounding medium. Experiments with new instruments and electronic solid state sensors are being used now to determine animal response to impending catastrophic occurrences.


The benefit from such research would be in duplicating the sensory responses of animals to construct equally responsive instruments that can record or monitor these precursory changes. Thus, observing and studying animal behavior could lead to better earthquake prediction instrumentation.

Operational Network


Since China considers such information on animal behavior vital to prediction, it established in 1968 its first experimental station for earthquake predictions making use of biological observations. This experimental station was established in Hsingtai Province. Other similar stations were set up in 1971 in Aksu, Sinkiang Province, where earthquakes were expected to occur. Since 1971, the Chinese have established an operational network in different communes or counties. Whenever unusual events occur and are reported by numerous observers, these are evaluated as a way of predicting earthquakes. So far, by this means, two major earthquakes have been predicted. This is easy for the Chinese since 80 percent of the population live in farming areas that are in close association with animals which can be observed readily. It is a little more difficult for people living in urban areas to observe similar animal behavior.