At 11:56 a.m. on July 4, those who were indoors in Beijing felt the earth shake like a boat in a mild storm. A few minutes later, they learned from radio, TV or the Internet that the epicenter of the magnitude 5.1 earthquake was in Wenan County, Hebei Province, 110 km from downtown Beijing.

The earthquake was as much of a surprise to seismologists as it was to ordinary citizens, who were rattled by the experience. "We failed to predict the quake in advance," said Huang Jianfa, a senior official from the China Earthquake Administration.

This is only one more example of the difficulty of forecasting earthquakes.

China is located in one of the most seismically active regions in the world and has been plagued by numerous destructive earthquakes throughout its long history. Over one third of the world's earthquakes strike China, which accounts for about 7 percent of the world's total land area. In the last century alone, nearly one third of the world's earthquakes above magnitude 7 took place in China.

Over the past thousand years, the Chinese people have been trying to find a way to predict earthquakes, but to little effect. However, seismologists are working to find better and more accurate forecasting methods.

Haicheng success

The most successful earthquake prediction case in China was over 30 years ago, one year before the disastrous Tangshan earthquake jolted the country.

On February 4, 1975, Haicheng City and surrounding areas in northeastern Liaoning Province were hit by a 7.3-magnitude earthquake, affecting 8 million people and an area of over 9,000 square km. As early as 1970, the State Seismological Bureau (known as the China Earthquake Administration today) had identified Liaoning as an area of high earthquake risk. Based on studies of precursor events and unusual animal behavior, in January 1975, state and local seismological bureaus made medium- and short-term predictions that a strong earthquake above magnitude 6 would strike areas involving Yingkou, Haicheng and Dalian.

After the announcement was made to the pubic, local governments at all levels adopted effective relief measures. From February 1, Haicheng was struck by small tremors with increasing frequency. On that date, the local seismological department predicted that there would be an immense earthquake in 24 hours, and reported their forecast to the local government immediately.

"Over 90 percent of the houses in quake-hit areas collapsed," said Hu Xiaohua, Vice Governor of Liaoning. "According to estimates, in a medium-sized city like Haicheng, without a forecast, over 100,000 inhabitants would have died in an above 7-magnitude earthquake. But fortunately, the majority of the millions affected had been evacuated to safe places beforehand. As a result, only 1,328 lives were claimed," he added.

Because of the successful forecast, the United Nations Educational, Scientific and Cultural Organization recognized China as the only country to have successfully predicted an impending quake.

"It is indeed the first effective and immediate earthquake prediction in human history," Chen Jianmin, head of the China Earthquake Administration, pointed out. Chen said this indicated that at least some types of earthquakes are predictable. "It brings hope for earthquake prediction and disaster relief. Meanwhile, it also boosts scientists' confidence in predicting earthquakes," Chen noted.

Before and after the Haicheng earthquake, governments at all levels in Liaoning took decisive measures to alleviate the destructive effects of the earthquake. Therefore, experts noted that the successful prediction set a precedent for an integrated earthquake relief method featuring scientific prediction, risk management and disaster relief.

Incessant efforts

The earthquakes that struck Xingtai City in Hebei on March 8 and 22, 1966, with respective magnitudes of 6.8 and 7.2, stimulated earthquake monitoring and prediction efforts in China. These two earthquakes destroyed over 5 million houses, claimed 8,064 lives and injured over 30,000 people. Farmland, roads and bridges were all ruined.

Late Premier Zhou Enlai called on scientists to sum up some hints for earthquake prediction and reporting from those two cases. Therefore, the Central Government mobilized 54 research institutes to form a 2,600-member integrated earthquake investigation and observation team. The team went to Xingtai to study the situation.

The scientists found that earthquakes tend to follow a pattern in which a number of temblors occur in a relatively short period, followed by a period of little or no activity. The Xingtai earthquake occurred during the fourth active period in China in the 20th century. This climax began in 1966 and ended in 1976 with the Tangshan magnitude 7.8 earthquake and the Songpan magnitude 7.2 earthquake in Sichuan Province. Over the approximately 10 years, 14 earthquakes above magnitude 7 struck the Chinese mainland.

The earthquakes caused severe damage, but at the same time provided scientists with unprecedented experience and conditions for studying and predicting earthquakes.

Organized by the government, a large-scale earthquake monitoring system had been formed on the Chinese mainland by the early 1990s. The network consists of over 400 national- or provincial-level monitoring stations, 30 regional surveillance networks and over 1,700 outposts. The mobile monitoring system extends over 10,000 km with nearly 10,000 amateur reporters.

Experts say the network, vast in its coverage, with multiple monitoring methods and an immense scale, is "unmatchable in the world." The network has not only provided abundant and precious basic materials for seismological studies and geosciences, but also laid a foundation for earthquake prediction.

On the basis of data collected from a large number of earthquakes and laboratory research, experts have made an initial exploration of the physical significance of the phase-to-phase warning signs of an earthquake, the phase-to-phase prediction value and procedures and grounds for making forecasts. China has gradually established an approach to earthquake prediction including long-, medium- and short-term monitoring, as well as monitoring right before an earthquake, which targets the overall earthquake situation, the zone likely to be struck by an earthquake and its intensity one or several years prior to the event, and the situation and disaster relief measures several days ahead of an earthquake.

After the efforts of the past 30 years, seismologists have boasted of initial ability to predict the exact time, place and intensity of an earthquake. "In the past 30 years, nearly 30 earthquakes have been predicted successfully to some extent," said Chen of the China Earthquake Administration.

Still hard to predict

"It must be noted that successful predictions, compared with the huge number of earthquakes that fail to be predicted, represent just a small number," said Luo Lange, chief engineer of seismology with the Hebei Earthquake Administration. He added that, in terms of earthquake prediction, China is at the same low level as other countries in the world. Cases of false predictions and no predictions are far more numerous than successful forecasts.

During the monitoring process, scientists try to find the power source that triggers the earthquake, whether it is underground or at ground level. However, a natural barrier preventing them from monitoring quakes is the earth's deep and thick crust. Temblors mainly happen in the earth's crust about 15 km or even deeper—sometimes dozens or hundreds of km below the surface. Even using the most advanced technology, the drilling depth is only 12 km. "We are still 'shallow' in terms of monitoring and detecting earthquakes," noted Chen. "We cannot even drill the earth's crust."

Furthermore, following a strong earthquake, there is usually a long period of time before another one strikes the same area. Additionally, even in an earthquake-prone country such as China, there is still a low probability of an earthquake. "It is very hard to predict an earthquake successfully and find effective prediction methods in a short period of time, due to the small probability of an earthquake and few chances to achieve the mission," Chen said.

Some people believe that peculiar weather or strange animal behavior will indicate an earthquake. However, this forecasting method is not very reliable.

"An earthquake is a sudden movement of the earth's crust caused by the release of stress accumulated along geologic faults or by volcanic activity," Chen explained. "During the process, a series of phenomena will occur, for instance, a change in geomagnetism, a fluctuation of chemistry, odor, ground temperature or the underground water table. These may in turn affect the animals' behavior. Take snakes for instance. Snakes hibernate in winter. However, if the temperature goes up for whatever reason, snakes will assume that spring has arrived and it is time to come out."

Animals often display unusual behavior, which, to a large extent, is caused by reasons other than earthquakes. Therefore, from scientists' perspective, it is nearly impossible to predict earthquakes merely by studying animal behavior.

There are many other difficulties and complications preventing successful earthquake prediction. Still, the public hopes that one day earthquake prediction will be as easy as a weather forecast. Scientists are confident that some day in the future, a brand new "earthquake forecast" will be shown on TV, although they are not sure when the day will come.

"With the improvement in science and our understanding of the physical world and nature, I believe that we can successfully predict earthquakes just as people require," Chen said.