During the late 1960s and early 1970s, the International Biological Program (IBP) began to study how water affects productivity in different ecosystems. It was not until this current group of scientists pooled long-term data in a workshop at the National Center for Ecological Analysis and Synthesis at University of California, Santa Barbara that the scope and similarity of productivity in all ecosystems was seen. Their study, published in the journal Nature, represents datasets from 14 ecosystems.
The RUEmax model predicts that when water is the most limiting resource there will be convergence in the production-precipitation ratio for all biomes. The previous, site model predicts there will be a difference in productivity and that sensitivity of different ecosystems is a function of the different forms of plant growth.
"We looked across nine different biomes - from a forest all the way to a very dry desert - at the sensitivity of production to precipitation," said Smith, associate professor of Ecology and Evolutionary Biology at Yale. "The data showed that if you look at the driest year, when water is most limiting, in any of the 14 sites - and compare growth production in that year - the rate of growth production in all biomes converges to a RUEmax that is very similar to what is seen in deserts."
Therefore, in a forest, during a really dry year there is the same growth production per unit of precipitation as there is in a really dry year in a desert. This data indicates that production sensitivity depends on the extent to which water is the limiting resource.
Although the RUEmax model is based on previously collected and pooled data, experiments in progress uphold this model. At one grassland site in Kansas where rainfall was excluded completely, the observed growth ratio matched RUEmax prediction - much lower than the site model prediction.
"These results have strong implications for the future. Current global models do not take into account climate variation much beyond what has already been seen," said Smith. "A danger is that if we have a sudden drought, according to the RUEmax data, the older global models are likely to overestimate the capacity for plant production. We can't underestimate the value for this kind of study on making real predictions."
"There will be always be adaptation to different climate, but fluctuations in the rate of change are most important to ecosystem sensitivity and immediate responses will be more drastic," said Smith.
If precipitation becomes more variable, the variability itself will increase the sensitivity and vulnerability of the ecosystem with the same impact as reducing the precipitation.
Other investigators included Travis Huxman of the University of Arizona who was co-author; Philip A. Fay of the Natural Resources Research Institute, Duluth, Minn.; Alan K. Knapp of Colorado State University in Fort Collins; M. Rebecca Shaw of the Carnegie Institution of Washington in Stanford, Calif.; Michael E. Loik of the University of California, Santa Cruz; Stanley D. Smith of the University of Nevada, Las Vegas; David T. Tissue and John C. Zak of Texas Tech University in Lubbock; Jake F. Weltzin of the University of Tennessee in Knoxville; William T. Pockman of the University of New Mexico in Albuquerque; Osvaldo E. Sala of the University of Buenos Aires in Argentina; Brent M. Haddad of the University of California, Santa Cruz; John Harte of the University of California, Berkeley; George W. Koch of Northern Arizona University in Flagstaff; Susan Schwinning of Biosphere 2 Center, Columbia University, Oracle, Ariz.; Eric E. Small of the University of Colorado at Boulder; and David G. Williams of the University of Wyoming in Laramie. The research was funded by the National Science Foundation, the U.S. Department of Agriculture, the U.S. Department of Energy and the National Park Service.
Citation: Nature 429: 651 - 654. (10 Jun 2004)