By CARL T. HALL
San Francisco Chronicle
November 17, 2005
Climate change experts led by Tim Barnett at the Scripps Institution of Oceanography in La Jolla, Calif., found that at least one-sixth of the world's population, including much of the industrial world and a quarter of global economic output, appeared vulnerable to water shortages brought about by climate change.
Details appear today in the journal Nature, along with a separate study suggesting climate models are proving to be an effective way of analyzing and forecasting disruptions in water supplies brought on by global warming.
Most experts see a clear warming trend over much of the world, although regional impacts may vary. All leading computer models of the global climate system indicate that natural variability isn't enough to explain the changes being observed, causing most observers to conclude that human activities, notably the emission of carbon and other heat-trapping greenhouse gases, are the culprit.
Earlier work by Barnett and others has documented the regional impact of climate change on California, much of which depends on seasonal snowmelt in the Sierra Nevada to keep water taps flowing and farmlands irrigated.
The latest study was an attempt to expand this sort of regional study to encompass the entire globe, by identifying areas most likely to feel the pinch of declining water supplies because of their reliance on glacial meltwater and snowmelt.
Barnett and his colleagues _ Jennifer Adam and Dennis Lettenmaier of the University of Washington _ excluded some areas, including watersheds of the Colorado River in the western United States and the Angara River in Asia, where reservoir storage capacity was judged large enough to "buffer large seasonal stream flow shifts."
Some heavily populated areas downstream of clearly runoff-dependent regions also were excluded _ even though they, too, would most likely suffer _ simply because the scientists lacked a reliable data source.
Despite this conservative approach, Barnett said in an interview, he was a bit taken aback by the extent of the world map falling within the climatic red zone of impending water difficulties.
"This shows a rather dramatic region, a surprisingly large part of the Earth, where you would expect to have serious water-supply problems in the next several decades," Barnett said.
The warming trend already is showing effects in California's Sierra Nevada snowpack, this region's main water source.
Climate models suggest average temperatures in the West will be about 1 to 3 degrees warmer by 2050 than at present. Even though total precipitation isn't expected to change by much, because of the higher temperatures more of it will come as rain rather than snow. At the same time, the spring runoff will come about one month earlier in the year.
Expanding populations, agricultural and industrial interests, and the need to keep streams flowing to protect vulnerable fish and other species all promise to make the water situation even worse as the climate shifts.
"I think this will be one of the first greenhouse gas-related problems that will fall on the civilized world," Barnett said.
Some parts of the world, including a broad swath of Asia and India, rely heavily on glacial runoff during summer months. That flow is expected to increase as the glaciers recede because of warming, but that just means the "water shortage, when it comes, will likely arrive much more abruptly, in time, with water systems going from plenty to want in perhaps a few decades or less," Barnett said.
All long-term climate projections are subject to attack from skeptics who either doubt the reliability of the computer models or caution against overreacting. The Bush administration and allied climate advisers have adopted a generally cautious approach, calling for more study of the problem.
Bonner Cohen, a senior fellow at the conservative National Center for Public Policy Research in Washington, D.C., said it would be only prudent for water planners in the zone Barnett identified to expand their storage capacity _ just in case.
"The one word of skepticism I have on these studies is that ultimately we are talking about modeling, and modeling just doesn't have a good track record for predicting the future," he said. "Basing public policy just on climate models can be a very, very risky business. I would be very dubious selecting one study, no matter how well peer-reviewed, predicting the climate 25, 50 or 100 years into the future, when there are so many factors involved in the climate that at this point are so poorly understood."
A separate study in Nature, by P.C.D. "Chris" Miller of the U.S. Geological Survey and colleagues, added some reassurances on that score, suggesting that "an ensemble" of 12 computer climate models all pointed in essentially the same troubling direction: less available water for a warming planet.
Submit A Letter to the Editor