"The Response of Global Terrestrial Ecosystems to Interannual Temperature Variability" was written by B.H. (Rob) Braswell, Ernst Linder, and Berrien Moore, all of the University of New Hampshire (UNH); and David Schimel, National Center for Atmospheric Research (NCAR) in Boulder. According to Schimel, the results highlight the power of new data sets on global change, as well as the usefulness of computer models that connect the atmosphere and biosphere. "We were looking specifically for delayed ecosystem responses in this study because they had been predicted by the models," Schimel notes.
Braswell conducted much of the analysis during a graduate fellowship in global change at NCAR sponsored by Oak Ridge Associated Universities and the University Corporation for Atmospheric Research (UCAR) Climate System Modeling Project. NCAR is operated by UCAR under sponsorship of the National Science Foundation.
The authors used three main sources of data for the period 1979- 94, each compiled and distributed with support from the U.S. Global Change Research Program:
--temperatures derived by satellite for the atmosphere's lowest few kilometers by a National Aeronautics and Space Administration (NASA) microwave sounding unit
--a vegetation index calculated with input from a space-based NOAA radiometer (radiation sensor)
--atmospheric carbon dioxide records collected at the South Pole and at Mauna Loa, Hawaii, by the National Oceanic and Atmospheric Administration (NOAA). Because carbon dioxide is well mixed globally on long time scales, these sites provide a good measure of year-to-year variability.
The global temperature record revealed several multiyear patterns, including warming associated with El Nino events in the 1980s. These patterns were correlated globally with CO2 levels and regionally with vegetation growth. Global carbon dioxide levels, which are steadily rising due to human activities, tended to rise more quickly over the first few months after a global temperature peak. The carbon dioxide levels then rose at a slower pace during the one-to-three-year period after the temperature peak, followed by a gradual reacceleration.
The authors studied the temperature-vegetation relationship by region at data points separated by one degree latitude and longitude (roughly 85 by 110 kilometers, or 50 by 70 miles, at midlatitudes). At the peak of a warm period, plant growth tended to increase in polar and temperate regions and decrease at lower latitudes, including tropical rain forests and drier savanna/grassland regimes. "This contrast suggests that . . . temperature may have direct negative impacts on plant growth or may increase water stress in semiarid ecosystems," the authors note.
However, in the one-to-three-year period after a temperature peak, the patterns appear to reverse: plant growth is enhanced in the warmer and drier regions and limited at higher latitudes. Thus, low-latitude plant growth appears to be driving the enhanced uptake of carbon dioxide during this period.
The paper highlights the importance of regional analyses of climate change to detect areas where effects may run counter to a global average. "This is the first data-based study to consider regionally specific ecosystem responses on a global scale," says Schimel. "The results show quantitatively that ecosystems are sensitive to temperature perturbations."
Writer: Bob Henson
Find this press release on the World Wide Web at http://www.ucar.edu/ucargen/press/contents.html
To receive UCAR and NCAR press releases by e-mail, telephone 303-497-8601 or e-mail butterwo@ucar.edu