Multiple-oxygen isotope constraints on aerobic respiration and gross primary productivity.
Primary production on land and in the surface ocean is a critical component of Earth's carbon and oxygen cycles, controlling the uptake of CO 2 and the release of O 2 to the atmosphere. Quantitative estimates of these O 2 and CO 2 fluxes remain challenging. Of the methods used, the triple-oxygen-isotope analysis of O 2 is considered to provide the most robust in situ estimate of gross productivity. However, this method relies on one key assumption-that the oxygen isotope fractionation of aerobic
Primary production on land and in the surface ocean is a critical component of Earth's carbon and oxygen cycles, controlling the uptake of CO 2 and the release of O 2 to the atmosphere. Quantitative estimates of these O 2 and CO 2 fluxes remain challenging. Of the methods used, the triple-oxygen-isotope analysis of O 2 is considered to provide the most robust in situ estimate of gross productivity. However, this method relies on one key assumption-that the oxygen isotope fractionation of aerobic respiration is constant-and recent experimental studies have shown that it can vary significantly. When applied to the same calculation of gross productivity, the experimentally determined variation in fractionation could produce [Formula: see text]100% error. In this work, the mechanistic origin of the variation in respiratory fractionation is explored using experiments on the model organism Escherichia coli . These experiments suggest that the fractionation is strongly correlated with the cell-specific O 2 consumption rate, weakly correlated with the dissolved-O 2 concentration, and may also depend on the identity of the terminal oxidase enzyme used to reduce O 2 . Modeling suggests that these influences on fractionation are likely to be important in the marine water column, and could produce up to [Formula: see text]120% error in estimates of gross productivity made using triple-oxygen-isotope analyses. This error may be significantly reduced through concurrent analyses of the "clumped" (multiply substituted) isotopologue abundances in O 2 .