MOMENTUM, WATER-VAPOR, AND CARBON-DIOXIDE EXCHANGE AT A CENTRALLY LOCATED PRAIRIE SITE DURING FIFE

Eddy correlation measurements were made of fluxes of momentum, sensible heat, water vapor, and carbon dioxide at a centrally located plateau site in the FIFE study area during the period from May to October 1987. About 82% of the vegetation at the site was comprised of several C4 grass species (big bluestem, Indian grass, switchgrass, tall dropseed, little bluestem, and blue grama), with the remainder being C3 grasses, sedges, forbs, and woody plants. The prairie was burned in mid-April and was not grazed. Precipitation during the study period was about normal, except for a 3-week dry period in late July to early August, which caused moisture stress conditions. The drag coefficient (Cd = u*2/u2BAR, where u* is the friction velocity and uBAR is the mean wind speed at 2.25 m above the ground) of the prairie vegetation ranged from 0.0087 to 0.0099. The average d/z(c) and z0/z(c) (where d is the zero plane displacement, z0 is the roughness parameter, and z(c) is the canopy height) were estimated to be about 0.71 and 0.028, respectively. Information was developed on the aerodynamic conductance (g(a)) in terms of mean wind speed (measured at a reference height) for different periods in the growing season. During the early and peak growth stages, with favorable soil moisture, the daily evapotranspiration (ET) rates ranged from 3.9 to 6.6 mm d-1. The ET rate during the dry period was between 2.9 and 3.8 mm d-1. The value of the Priestley-Taylor coefficient (alpha), calculated as the ratio of the measured ET to the equilibrium ET, averaged around 1.26 when the canopy stomatal resistance (r(c)) was less than 100 s m-1. When r(c) increased above 100 s m-1, alpha decreased rapidly. The atmospheric CO2 flux data (eddy correlation) were used, in conjunction with estimated soil CO2 flux, to evaluate canopy photosynthesis (P(c)). The dependence of P(c) on photosynthetically active radiation (K(PAR)), vapor pressure deficit, and soil moisture was examined. Under nonlimiting soil moisture conditions, P(c) was primarily controlled by K(PAR) through a rectangular hyperbolic relationship. Our data did not indicate light saturation of the canopy up to K(PAR) levels of 2100 muEi m-2 s-1. Midday values of P(c) reached a seasonal peak of 1.4-1.5 mg m 2 (ground area) s-1 during late June and early July. During the dry period (late July to early August), midday P(c) declined to a minimum of almost zero. Examination of data on P(c), lambdaE/R(n) (the proportion of net radiation consumed in latent heat flux), extractable soil water, and the predawn leaf water potential indicated a remarkable similarity in overall patterns throughout the season. The photosynthetic efficiency was 1.5-2% during midday through most of the growing season (except during the dry period). The midday value of the prairie water use efficiency during the peak growth stage was 8-12 x 10(-3) g CO2/g H2O.