<p>The O<sub>2 </sub>: CO<sub>2</sub> exchange ratio (ER) between terrestrial ecosystems and the atmosphere is a key parameter for partitioning global ocean and land carbon fluxes. The long-term terrestrial ER is considered to be close to 1.10 moles of O<sub>2</sub> consumed per mole of CO<sub>2</sub> produced. Due to the technical challenge in measuring directly the ER of entire terrestrial ecosystems (ER<sub>eco</sub>), little is known about the variations in ER at the hourly and seasonal scales as well as how different components contribute to ER<sub>eco</sub>. In this modeling study, we explore the variability and drivers of ER<sub>eco</sub> and evaluate the hypothetical uncertainty in determining ecosystem O<sub>2</sub> fluxes based on current instrument precision. We adapted the one-dimensional, multi-layer atmosphere-biosphere gas exchange model, CANVEG, to simulate hourly ER<sub>eco</sub> from modeled O<sub>2</sub> and CO<sub>2</sub> fluxes in a temperate beech forest in Germany.</p> <p>We found that the annual mean ER<sub>eco</sub> ranged from 1.06 to 1.12 mol mol<sup>-1</sup> within the five years’ study period. Hourly ER<sub>eco</sub> showed strong variations over diel and seasonal cycles and within the vertical canopy profile. Determination of ER from O<sub>2</sub> and CO<sub>2</sub> mole fractions in air above and within the canopy (ER<sub>conc</sub>) varied between 1.115 and 1.15 mol mol<sup>-1</sup>. CANVEG simulations also indicated that ecosystem O<sub>2</sub> fluxes could be derived using the flux-gradient method in combination with measurements of vertical scalar gradients and CO<sub>2</sub>, sensible heat or latent heat fluxes obtained with the eddy covariance technique. Owing to measurement uncertainties, however, the uncertainty in estimated O<sub>2</sub> fluxes derived with the flux-gradient approach could be as high as 15 μmol m<sup>-2</sup> s<sup>-1</sup>, which represented the 90 % quantile of the uncertainty in hourly data with a high-accuracy instrument. We also demonstrated that O<sub>2</sub> fluxes can be used to partition net CO<sub>2</sub> exchange fluxes into their component fluxes of photosynthesis and respiration, if ER<sub>eco</sub> is known. The uncertainty of the partitioned gross assimilation ranged from 1.43 to 4.88 μmol m<sup>-2</sup> s<sup>-1</sup> assuming a measurement uncertainty of 0.1 or 2.5 μmol m<sup>-2</sup> s<sup>-1</sup> for net ecosystem CO<sub>2</sub> exchange and from 0.1 to 15 μmol m<sup>-2</sup> s<sup>-1</sup> for net ecosystem O<sub>2</sub> exchange, respectively. Our analysis suggests that O<sub>2</sub> measurements at ecosystem scale have the potential for partitioning net CO<sub>2</sub> fluxes into their component fluxes, but further improvement in instrument precision is needed.</p>