FLUXES OF CO2 AND WATER-VAPOR FROM A PRAIRIE ECOSYSTEM EXPOSED TO AMBIENT AND ELEVATED ATMOSPHERIC CO2

Increasing concentrations of atmospheric CO2 may alter the carbon and water relations of prairie ecosystems. A C-4-dominated tallgrass prair ie near Manhattan, KS, was exposed to 2x ambient CO2 concentrations us ing 4.5 m-diameter open-top chambers. Whole-chamber net CO2 exchange ( NCE) and evapotranspiration (ET) were continuously monitored in CO2-en riched and ambient (no enrichment) plots over a 34-d period encompassi ng the time of peak biomass in July and August, 1993. Soil-surface CO2 fluxes were measured with a portable surface chamber, and sap flow (w ater transport in xylem) in individual grass culms was monitored with heat balance techniques. Environmental measurements were used to deter mine the effect of CO2 on the surface energy balance and canopy resist ances to vapor flux. In 1993, frequent rainfall kept soil water near f ield capacity and minimized plant water stress. Over the 34-d measurem ent period, average daily NCE (canopy photosynthesis - soil and canopy respiration) was 9.3 g CO2 m(-2) in the ambient treatment and 11.4 g CO2 m(-2) under CO2 enrichment. However, differences in NCE were cause d mainly by delayed senescence in the CO2-enriched plots at the end of the growing season. At earlier stages of growth, elevated CO2 had no effect on NCE. Soil-surface CO2 fluxes typically ranged from 0.4 to 0. 66 mg CO2 m(-2) s(-1), but were slightly greater in the CO2-enriched c hambers. CO2 enrichment reduced daily ET by 22%, reduced sap flow by 1 8%, and increased canopy resistance to vapor flux by 24 s m(-1). Great er NCE and lower ET resulted in higher daytime water use efficiency (W UE) under CO2 enrichment vs. ambient (9.84 vs. 7.26 g CO2 kg(-1) H2O). However, record high precipitation during the 1993 season moderated t he effect of WUE on plant growth, and elevated CO2 had no effect on pe ak aboveground biomass. CO2-induced stomatal closure also affected the energy balance of the surface by reducing latent heat flux (LE), ther eby causing a consequent change in sensible heat flux (H). The daytime Bowen ratio (H/LE) for the study period was near zero for the ambient treatment and 0.21 under CO2 enrichment.