Abstract Accurate representation of CO 2 transport is important for reliable flux inversions with global inversion systems. Previous studies have revealed systematic discrepancies between two widely used global transport models, GEOS‐Chem and TM5, in simulated CO 2 distributions at mid‐latitudes. However, the causes of these discrepancies and potential transport biases in the models remain unclear without constraints of intensive measurements. Here, we drive GEOS‐Chem and TM5 with identical surface CO 2 fluxes and perform simulations with and without parameterized convection. By comparing the simulated spatial‐temporal distributions of CO 2 to intensive aircraft and tower measurements collected by the Atmospheric Carbon Transport (ACT)‐America project and NOAA's airborne profiling and tower networks across four seasons, we identify potential transport biases in both models over North America. Model discrepancies in vertical transport produce seasonal‐mean planetary boundary layer (PBL)–free‐troposphere CO 2 differences of up to ∼4 ppm and PBL height differences of ∼100–400 m. In summer, GEOS‐Chem underestimates transport through moist convection but overestimates PBL mixing depth, whereas TM5 exhibits more realistic total vertical mixing (moist convection and PBL depth). In spring and winter, TM5 underestimates PBL mixing while potentially overestimating mixing through moist convection, whereas GEOS‐Chem more accurately represents vertical mixing. These transport biases are likely to contribute to discrepancies in the meridional gradient of zonal‐mean CO 2 and propagate into differences in posterior flux estimates from the OCO‐2 Model Intercomparison Project inversions using the two transport models over North America. Expanded vertically resolved CO 2 measurements beyond North America can further advance the current understanding of transport uncertainties at larger spatial scales.
Zhang et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: