Increase In Rainfall Variability Related To Global Climate Change; Impacts On Ecosystems Are Greater Than Previously Anticipated

Results of the experiment, conducted at NSF's Konza Prairie Long Term Ecological Research (LTER) site, are published in this week's (December 13th) issue of the journal Science.

The biologists, Alan Knapp, Philip Fay, and John Blair and colleagues of Kansas State University, Scott Collins of NSF, and Melinda Smith at the National Center for Ecological Analysis and Synthesis at the University of California, Santa Barbara, found that more extreme swings in rainfall patterns, without any changes in the total amount of rainfall received in a growing season, reduced the biomass of plants but increased the variety of species able to live in a particular experimental plot of land.

"This study is the first to focus on and manipulate climate variability in an intact ecosystem, without altering the average climate," said Quentin Wheeler, director of NSF's division of environmental biology, which funded the research along with the U.S. Department of Agriculture and the U.S. Department of Energy. "Because these responses are similar to those that would occur under drought conditions, the results suggest that increased rainfall variability combined with projected higher temperatures and decreased rainfall amounts, may lead to even greater impacts on ecosystems than previously anticipated."

In this study of how grasslands respond to more variation in rainfall patterns, the scientists hoped to better understand how rapidly and to what extent ecosystems might respond to a future with more climate extremes. In the four-year field study, the researchers altered rainfall variability by increasing the amount of precipitation that falls in one storm, and lengthened the periods of time between rainfalls by 50 percent. That effectively increased the severity of dry periods between storms without altering the total amount of precipitation received during a growing season.

"When these native grassland plots, exposed to more variable rainfall patterns, were compared with plots that received rainfall in a natural pattern, the overall growth of all plants decreased," said Knapp. "More variable rainfall patterns led to lower amounts of water in the soil in the upper 30 centimeters. Since this is the soil depth where most plant roots occur, and where important soil microbes are most abundant, grasses there were water-stressed and the activity of below-ground organisms was reduced."

In contrast, said Collins, "the diversity of plants in plots with greater variability in rainfall patterns increased." Collins cites two possible explanations for this finding: "A high degree of variability in resources can lead to a greater number of co-existing species. Or reduced total productivity may have allowed less common species to increase in abundance." Regardless of the mechanism, said Collins, these results show that plant community structure can be significantly changed, and the cycling of carbon slowed, in as little as four years when grasslands are exposed to a more variable climate.

Concerns about predicted climate changes resulting from human activities often focus on the effects of increases in average air temperatures or changes in average precipitation amounts. But widely used climate models also predict increases in climate extremes, said Knapp, such as more frequent large rainfall events or more severe droughts. "It's important that we look at variability in a new way: not only from year to year or decade to decade, but from storm to storm."