Eddy covariance (EC) is considered the most direct method for measuring fluxes of mass and energy between the surface and the atmosphere. Data from EC systems are often integrated over time to determine the seasonal and annual fluxes of carbon and water. Unfortunately, small biases in these data can lead to substantial errors in integrated fluxes. Recent data from several sites suggests that certain EC systems may underestimate ecosystem respiration during periods of low flux. Open- and closed-path EC measurements were made on a tallgrass prairie near Manhattan, KS. Automated surface chambers were used to measure ecosystem respiration in the source area of the EC tower. In theory, all three measurement systems should have given the same estimate of ecosystem respiration when no photosynthesis was occurring (i.e, at night or during the dormant season). Data from these systems will be compared under a wide range of flux rates and atmospheric conditions. Error analyses will be utilized in conjunction with the results to determine the expected accuracy of EC measurements and to what extent data can be integrated to obtain seasonal or annual flux estimates.