Interactions between water availability and elevated atmospheric CO2 concentrations have the potential to be important factors in determining future forage supply from temperate pastures. Using large turves from an established pasture, the response of these communities at 350 or 700 µl l−1 CO2 to a soil moisture deficit and to recovery from the deficit in comparison to turves that were well-watered throughout was measured. Prior to this experiment the turves had been exposed to the CO2 treatments for 324 d.

Net CO2 exchange continued at elevated CO2 even when the volumetric soil moisture content was less than 0.10 m3 m−3 soil; at the same moisture deficit gas exchange at ambient CO2 was zero. The additional carbon fixed by the elevated CO2 turves was primarily allocated below-ground as shown by the maintenance of root length density at the same level as in well-watered turves. When the dry turves were rewatered there was compensatory growth at ambient CO2 so that the above-ground growth rate exceeded that of turves that had not experienced a moisture deficit. At the start of this experiment, the turves that were growing at 700 µl I−1 CO2 had a greater proportion of legume (principally white clover, Trifolium repens L.) in the harvested herbage. There was a trend for the legume content at elevated CO2 to be reduced under a soil moisture deficit.

The results indicate different strategies in response to soil moisture deficits depending on the CO2 concentration. At ambient CO2, growth stopped, but plants were able to respond strongly on rewatering; while at elevated CO2 growth continued (particularly below-ground), but no additional growth was evident on rewatering. Ecosystem gas exchange measurements taken at the end of the experiment (after 429 d of exposure to CO2) showed 33% more CO2 was fixed at elevated CO2 with only a small (12%) and nonsignificant downward regulation.