C4 plants contribute ≈ 20% of global gross primary productivity, and uncertainties regarding their responses to rising atmospheric CO2 concentrations may limit predictions of future global change impacts on C4-dominated ecosystems. These uncertainties have not yet been considered rigorously due to expectations of C4 low responsiveness based on photosynthetic theory and early experiments. We carried out a literature review (1980–97) and meta-analysis in order to identify emerging patterns of C4 grass responses to elevated CO2, as compared with those of C3 grasses. The focus was on nondomesticated Poaceae alone, to the exclusion of C4 dicotyledonous and C4 crop species. This provides a clear test, controlled for genotypic variability at family level, of differences between the CO2-responsiveness of these functional types. Eleven responses were considered, ranging from physiological behaviour at the leaf level to carbon allocation patterns at the whole plant level. Results were also assessed in the context of environmental stress conditions (light, temperature, water and nutrient stress), and experimental growing conditions (pot size, experimental duration and fumigation method).
Both C4 and C3 species increased total biomass significantly in elevated CO2, by 33% and 44%, respectively. Differing tendencies between types in shoot structural response were revealed: C3 species showed a greater increase in tillering, whereas C4 species showed a greater increase in leaf area in elevated CO2. At the leaf level, significant stomatal closure and increased leaf water use efficiency were confirmed in both types, and higher carbon assimilation rates were found in both C3 and C4 species (33% and 25%, respectively). Environmental stress did not alter the C4 CO2-response, except for the loss of a significant positive CO2-response for above-ground biomass and leaf area under water stress. In C3 species, stimulation of carbon assimilation rate was reduced by stress (overall), and nutrient stress tended to reduce the mean biomass response to elevated CO2. Leaf carbohydrate status increased and leaf nitrogen concentration decreased significantly in elevated CO2 only in C3 species.
We conclude that the relative responses of the C4 and C3 photosynthetic types to elevated CO2 concur only to some extent with expectations based on photosynthetic theory. The significant positive responses of C4 grass species at both the leaf and the whole plant level demand a re-evaluation of the assumption of low responsiveness in C4 plants at both levels, and not only with regard to water relations. The combined shoot structural and water use efficiency responses of these functional types will have consequential implications for the water balance of important catchments and range-lands throughout the world, especially in semiarid subtropical and temperate regions. It may be premature to predict that C4 grass species will lose their competitive advantage over C3 grass species in elevated CO2.