- Top of page
- Materials and Methods
The growth and yield of plants increases when the concentration of atmospheric CO2 is doubled (Kimball, 1983; Cure & Acock, 1986), in particular at high nutrient availability (Kimball, 1993; Poorter et al., 1996; Curtis & Wang, 1998). Growth responses to elevated CO2 vary greatly among species (Poorter, 1993). Crop plants and fast-growing wild species increase their biomass to a greater extent than slow-growing wild species (Hunt et al., 1991, 1993); legumes, with their unlimited access to the N2 pool in the atmosphere (symbiotic N2 fixation), usually show a stronger response to elevated CO2 than nonfixing plant species (Newton et al., 1994; Lüscher et al., 1996; Soussana & Hartwig, 1996; Clark et al., 1997; Hebeisen et al., 1997b; Lüscher et al., 2000). If nutrients other than N (e.g. P) are strongly limited, however, the advantage of the legumes disappeared (Stöcklin et al., 1998; Almeida et al., 1999, 2000). These results suggest that a high growth rate and a high nutrient availability promote a strong response to elevated CO2.
In dense swards under field conditions, the response of plants to elevated CO2 may be strongly affected by interactions among plants (e.g. competition for limited resources such as light and nutrients) and by interactions between plants and the soil (e.g. nutrient cycling). Perennial ryegrass (Lolium perenne), a fast growing species that requires a large amount of nitrogen (N), is the most important species in intensively managed temperate grasslands. Several studies were conducted to examine the effect of elevated CO2 on L. perenne (Newton et al., 1994; Casella et al., 1996; Hebeisen et al., 1997b; Daepp et al., 2000), but the results varied with respect to the yield increase at elevated CO2. In the first 2 yr of the Swiss FACE experiment, elevated CO2 had a weak effect on the DM yield of L. perenne swards (Hebeisen et al., 1997b; Daepp et al., 2000) and of other grass species (Lüscher et al., 1998). At the same time, the N concentration, N yield, specific leaf area and the N nutrition index of L. perenne were lower at elevated CO2 (Soussana et al., 1996; Hebeisen et al., 1997a; Zanetti et al., 1997). Under the same conditions, however, the legumes showed no signs of N limitation, and symbiotic N2 fixation increased under elevated CO2 (Zanetti et al., 1996, 1997). From these results, it was suggested that an inadequate supply of mineral N in the soil limited the response of L. perenne to elevated CO2 under field conditions, even in the high N treatment (56 g m−2 y−1) (Hebeisen et al., 1997b). The yield response of L. perenne swards to elevated CO2, however, increased over the six years of the FACE experiment. This increase was attributed to a steady increase in the availability of mineral N in the soil (Daepp et al., 2000). However, effects of the age of the sward and of climatic conditions in the different years could not fully be excluded. Thus, it remains to be determined whether N availability is the driving force of the response of L. perenne to elevated CO2 in dense swards and whether the high N treatment (56 g m−2 y−1) still limits for the growth and the response to CO2.
In the Swiss FACE, the concentration and the night-time export of water soluble carbohydrate indicated that the yield response of L. perenne swards to elevated CO2 during vegetative growth was restricted by a C-sink limitation, which was particularly severe at low N supply (Fischer et al., 1997; Rogers et al., 1998; Isopp et al., 2000a,b). Fast growth at high N fertilization partly reduced the C-sink limitation. Not only nutrient availability, but also the growth stage may cause considerable variation in the strength of the C-sink. Reproductive growth of temperate grasses during the first growth cycle in spring triggers rapid stem elongation and therefore changes canopy structure. The rapidly expanding C-sinks may enable plants to overcome C-sink limitation of the response to elevated CO2 (Long & Hutchin, 1991). In fact, reproductive growth is characterized by a yield production rate that is 2–3 times greater than during vegetative growth (Menzi et al., 1991). Thus, comparing the response of vegetative and reproductive swards of L. perenne with elevated CO2 helps to clarify the effect of C-sink for the response to elevated CO2.
By contrast to the low DM yield response in the Swiss FACE (Hebeisen et al., 1997b), L. perenne showed a strong and consistent increase in photosynthetic C fixation per unit leaf area, independent of N fertilization and point of time in the regrowth cycle (Rogers et al., 1998; Isopp et al., 2000b). This indicates that C assimilation and DM yield are not closely related and that the allocation of carbohydrates to shoots and roots may change at elevated CO2. L. perenne showed a substantial increase in root biomass at elevated CO2 when measured at the end of the growing season (Hebeisen et al., 1997b). At elevated CO2, the root : shoot ratio of plants was often higher although this was not a consistent finding (Stulen & den Hertog, 1993; Luo et al., 1994). To understand the changes that occur in managed grassland ecosystems at elevated CO2, it is necessary to consider not only the yield, but also the residual biomass (roots and stubble). Data from field experiments, however, are sparse since the destructive harvesting of roots and stubble destroys the experimental plots. Thus, model swards were established in long pots (0.6 m) and could be harvested destructively over 2 yr without disturbing the remaining sward.
The objective of this 2-yr field experiment with model swards of L. perenne was to investigate the combined effects of elevated CO2 (600 ppm), N availability and the developmental stage on yield and allocation of DM. We hypothesized that, when L. perenne swards grow fast as a result of excessive N fertilization or reproductive growth, they have the potential to increase their yield under elevated CO2. Elevated CO2 is expected to increase root : shoot ratio except under excessive N fertilization.