Comparing studies of artificial and natural ecosystem responses to CO2 enrichment.

Abstract

Can experimental investigations of complex systems provide a clearer basis on which to make predictions about the real world? Can model ecosystems predict, at least qualitatively, natural effects? Using two superficially very different grassland ecosystems (natural alpine grassland and grassland model ecosystems) the effects of elevated atmospheric carbon dioxide (CO_2) on these systems are compared. Although very different in composition, climatic conditions and successional stage, results from the two experiments show considerable similarities. Both experiments also highlight the implications for long-term feedback processes in soil ecosystems that are subject to rising global atmospheric CO_2 concentrations. Finally, while studies show that small-scale, short-term single species experiments can yield misleading results we advocate the use of either field manipulations of real ecosystems or, if this is not possible for reasons of cost or time, more realistic (and complex) model systems (microcosms).