Measurements of net ecosystem CO2 exchange (NEE) and energy balance were made using chamber-, tower-, and aircraft-based measurement techniques in Alaskan arctic tundra ecosystems during the 1994--1995 growing seasons (June--August). One of our objectives was to quantify the interrelationships between the NEE and the energy balance measurements made from different sampling techniques. Qualitative and quantitative intercomparisons revealed that on average the correspondence between the mass and energy fluxes measured by these sampling methods was good despite potential spatial and temporal mismatches in sampling scale. Quantitative comparisons using least squares linear regression analyses with the tower-based measurements of NEE as the independent variable indicate that the chamber- and aircraft-based NEE measurements were generally lower relative to the tower-based measurements (slope=0.76--0.86). Similarly, tower-aircraft comparisons of latent (Le) and sensible (H) heat exchange indicated that the aircraft-based measurements were lower than the tower-based measurements (slope=0.72--0.80). Qualitative comparisons, however, indicate that the correspondence among the chamber-, tower-, and aircraft-measured fluxes varied both seasonally and interannually, suggesting the lack of a consistent bias between the sampling techniques. The results suggest that differences observed between the chamber, tower, and aircraft flux measurements were primarily due to the failure to account for the spatial distribution of surface types in the tower and aircraft sampling footprint, problems involved in the comparison of temporal and spatial averages, and temporal (e.g., seasonal and interannual) variance in rates of mass and energy flux for a given point. Other potential sources of variance include the underestimation of nocturnal NEE by the tower-based eddy covariance system, and the periodic occurrence of an elevated CO2 plume in the atmosphere over the Prudhoe Bay oil field. Even with these potential sources of variation, the results reveal that the various methods give comparable estimates of NEE and energy flux within a range of temporal or spatial variability. ¿ 1998 American Geophysical Union