Dry matter yield, leaf growth and population dynamics in Lolium perenne Trifolium repens-dominated pasture turves exposed to two levels of elevated CO2

1. Dry matter yield, leaf growth and population dynamics of turves taken from an old Lolium perenne/Trifolium repens-dominated pasture were studied in controlled environment rooms at CO2 concentrations of 350 mu mol mol(-1), 525 mu mol mol(-1) and 700 mu mol mol(-1). Starting with September data the turves were subjected sequentially to the mean monthly temperature and photoperiod taken from long-term climatic data for the area of New Zealand where the pasture was located. Each temperature and photoperiod regime was applied for 21 days to provide 12 different simulated 'months' of environmental conditions. The experiment ran for 14 simulated months, with September and October conditions being repeated at the end of the first simulated 'year'. Mean photon flux density throughout was 500 mu E m(-2) s(-1). 2. The total quantity of herbage harvested was increased by 7% and 14% at 525 mu mol mol(-1) and 700 mu mol mol(-1) CO2, respectively. The increase in the amount of T. repens harvested by the end of the experiment was 63% at 525 mu mol mol(-1) CO2 and 48% at 700 mu mol mol(-1) CO2. In contrast, neither the yield of C3 grasses nor the yield of the only C4 grass present, Paspalum dilatatum, was significantly affected by CO2 concentration. The implications of this increase in the proportion of T. repens in temperate pastures at elevated CO2 is discussed briefly. 3. With the exception of a small increase in the specific leaf area of T. repens, detailed measurements of leaf growth on marked tillers (L. perenne and P. dilatatum) and growing points (T. repens) showed no consistent effects of exposure to elevated CO2 concentrations. 4. Differences in yield between CO2 concentrations were mainly attributable to changes in the number and balance of population units. By the middle of the 'winter' conditions T. repens growing point densities at 700 mu mol mol(-1) CO2 were more than double those found at 350 mu mol mol(-1) but total grass tiller densities were unchanged. Growing point densities were also more than doubled at 525 mu mol mol(-1) CO2 compared with 350 mu mol mol(-1) but grass tiller densities were reduced significantly below those recorded in the other two treatments. The relationship between the stability of herbage production and population densities is discussed and the potential interaction between population density, elevated CO2 and grazing considered. 5. Although exposure to elevated levels of CO2 did result in large changes in population numbers, this did not happen immediately and so the yield response of this particular community to CO2 varied with time. The average yield increases recorded here at elevated CO2 may therefore tend to underestimate those likely to be shown by communities that, at the population level, have become fully adapted to growth in a CO2-enriched atmosphere.