Copyright 1997 SCIENCE, the journal of the American Association for the Advancement of Science

Reporters covering severe floods in the United States occasionally make some mention of a possible link to global warming. As it turns out, however, flood patterns can be tracked over time and an assessment made as to whether they have indeed become more frequent. Over the last century, global temperatures have warmed approximately 0.5 degrees celsius (1 degree fahrenheit), most of it occurring before 1940 as the earth recovered from the 400-year "Little Ice Age," which ended around 1850. Are we seeing more rain and more flooding from this modest rise in temperature? Two researchers presented their findings at the December 1997 conference of the American Geophysical Union, held in San Francisco, California. Here are their conclusions, as reported in Science magazine on-line. (www.apnet.com/inscight/12091997/grapha.htm).

SAN FRANCISCO, DECEMBER 9, 1997---Dramatic news coverage of deluged towns might suggest that floods in the United States have become larger and more frequent over the past few years--a trend that could be an early sign of global warming. But that perception is all wet, according to reports presented here yesterday at the fall meeting of the American Geophysical Union.

This year's devastating floods in the upper Great Plains and California's Central Valley may have bolstered the sense that watery disasters are on the rise. These severe events took 98 lives and caused some $8.7 billion in damage. Other notable floods have ravaged the country in this decade, including 1993's widespread inundation along the Mississippi River. Even so, most flooding experts saw no abnormal climatic patterns emerging. Now, detailed records from river gauges support that view, according to hydrologist Harry Lins and mathematician James Slack of the U.S. Geological Survey.

Lins and Slack scrutinized records of average water-flow rates and peak daily discharges from about 300 gauges scattered around the country, dating back as far as 1914. A statistical analysis uncovered no increase in either the frequency of floods or their size for the country as a whole. Although the records extend only through the mid-1990s, Lins believes the events of 1997 fall into the typical pattern and would not have altered their results. "Those were important and major floods, but they were confined to very specific regions," he says. "We do not see any evidence of a change in large-scale national patterns."

Other experts agree. "I don't think we can say that the climate has changed substantially, if at all," says meteorologist Frank Richards of the National Weather Service. What may have changed, Lins and Slack note, is development along flood-prone rivers. Because levees and structures may exacerbate property damage during certain floods, they only studied records from natural watersheds.