Scientists say the Himalayan mountain region is long overdue for one or more major earthquakes that could threaten millions of people in northern India and neighboring south Asian countries. The predicted loss of life would make it one of the worst earthquake death tolls.

The high Himalayas are natural testimony to the violence that occurs when Earth's rocky surface plates crash into each other. The Indian subcontinent was once separate from Asia, but the two land masses gradually came together over the eons, pushing up their edges into the world's tallest peaks.

The Himalayas cool serenity hides the continuing activity of India's basement rock, which flexes and dives slowly downward. Satellite measurements show that the Indian surface slips under the mountains about two centimeters a year. This adds up to two meters a century, with increasing pressure of one surface against the other.

The constant grinding causes many small earthquakes. But University of Colorado geophysicist Roger Bilham says most of the Himalayas have not had a major tremor for 300 years and one is necessary to release the enormous pressure buildup.

"We think there are approximately 10 regions where these giant earthquakes occur. Looking back into recent history, we find that only three of them have slipped, and we know that the others have to go," Mr. Bilham says. "For example, there was one in 1934 that released four meters of slip. This was a magnitude 8.1 earthquake. It appears that at least 70 percent of the Himalayas is overdue for that amount of slip."

The last major Himalayan earthquake occurred in Assam in 1950. At magnitude 8.5, it was the most powerful ever recorded on land. In an analysis in the journal Science, Mr. Bilham and colleagues at the Indian Institute for Astrophysics in Bangalore say another quake like it would be widely devastating because the population of the vulnerable Ganges plain just south of the mountains has grown 10 times in the past century. Mr. Bilham says strong shaking would imperil New Delhi and the capital cities of neighboring Bangladesh, Bhutan, Nepal, and Pakistan.

How many deaths might occur? A big Himalayan quake in 1905 killed nearly 20,000 people. With 10 times more people in the region today, the researchers predict at least 200,000 would die. "It's an alarming number," he says. "It could be less than that, but it could also be much more. One of the disturbing findings is if we had an earthquake near one of the larger populations centers like Lucknow or Delhi, the population at risk could be much higher."

Since 1803, there have been six Himalayan earthquakes of the magnitude of which Mr. Bilham speaks from seven to 64 years apart. Fifty-one years have passed since the last one, but earthquake expert Eugene Schweig of the U.S. Geological Service reminds us that predicting the timing of the next is impossible. "We know that there have been very large earthquakes in the past, so there is every reason to think that that possibility exists for the future," Mr. Schweig says. "But we don't think that we have the capability to predict earthquakes right now, whether they'd happen soon or 100 years from now."

What worries Roger Bilham is the susceptibility of the region's buildings. He says building codes intended to strengthen structures have not reduced the percentage of people killed in a major tremor. The 19,000 confirmed deaths from January's earthquake in Bhuj, which is not near the Himalayas, is the same percentage that occurred in a quake in that region in 1819.

Mr. Bilham says cheating by building contractors is prevalent. "Given the severity of these future events," he says, "I think just one more event will certainly prompt very, very serious enforcement of building codes in northern India."

Images Courtesy of size=1>