The story of some violent historic earthquakes may need to be revisited, according to a study published in the April issue of the Bulletin of the Seismological Society of America (BSSA). Seismologists rely on written accounts, mostly local newspaper articles, to judge how strongly the ground shook during earthquakes that predate the use of current instrumentation. Those news accounts have proven to be misleading, say scientists, and reliance upon them must be tempered when evaluating the size of past earthquakes.

By focusing on the most dramatic damage and other effects of an earthquake, news stories can provide an unbalanced view of a disaster. For historical earthquakes it is difficult to estimate the effects of this bias. However, a recently deadly earthquake--the M7.6 Bhuj, India earthquake of 2001--provided an unprecedented opportunity to compare the media accounts with the results of an exhaustive, ground-based survey of damage.

"This study isn’t about the media," says Susan E. Hough, co-author of the paper and a seismologist at the U. S. Geological Survey in Pasadena, California. "It isn’t the job of the media to provide a detailed survey of the effects of an earthquake. It’s the seismologist’s job to evaluate media and other written accounts. We need to do careful, balanced assessments of accounts of past earthquakes to understand the hazard from future earthquakes. Media accounts have a built-in bias that is natural to telling any story – whether by a journalist or by an eyewitness."

The severity of an earthquake can be expressed in terms of intensity, which is based on observations by eyewitnesses to the ground motion and the shaking of buildings, trees, and anything in the surrounding area. The perceived intensity varies within the affected zone, depending upon the location of the witness to the epicenter.

In "Quantifying the ‘Media Bias" in Intensity Surveys: Lessons from the 2001 Bhuj, India, Earthquake," Hough and co-author Prabhas Pande, Ph.D, Director, Earthquake Geology Division Geological Survey of India, studied the effects of the 7.6 Bhuj earthquake that occurred on 26 January 2001. This quake was felt across much of the Indian subcontinent, and official government figures cited 13,800 fatalities and 166,000 injuries.

Two independent intensity surveys evaluated the earthquake: one based on extensive news articles written in the early aftermath of the earthquake and the other based on direct surveys and interviews conducted by the Geological Survey of India. The comparison yielded important information for seismologists who interpret past events for which only written accounts are available.

"The research confirmed the tendency of written accounts to focus on the most dramatic, rather than representative effects in their accounts," write the authors. Further, the authors conclude that the media bias "can be significant, and is most severe at the strongest shaking levels."

Written from a news release by Seismological Society of America.