Atmospheric carbon monoxide (CO) is an important primary pollutant in the atmosphere. As a result of its relatively long lifetime, it is a good tracer of atmospheric transport. It is also of interest in a climate change context, due to the indirect radiative forcing effect on the climate system. The CO state and emission estimates both have large uncertainties, due to the uncertain OH budget and model errors. To improve our understanding, the four-dimensional variational (4D-Var) scheme has been conventionally used in data assimilation, in which the model is assumed to be perfect and used as a strong constraint (SC) to optimize initial/boundary conditions. However, models are imperfect, and the model errors have been shown to have an impact on the quality of data assimilation. In this project, we conduct a series of Observation System Simulation Experiments (OSSEs) on the GEOS-Chem chemical transport model (CTM) to examine the utility of the weak-constraint (WC) 4D-Var approach as a means of mitigating model transport biases in atmospheric data assimilation systems.