An assessment of the utility of CO2 vertical profile measurements by a solar occultation based satellite sensor is made. We have compared the impacts of these possible vertical profile observations and optimal extensions of the present surface measurement network on the estimation of CO2 regional sources by inverse model. Time-independent inverse models at two spatial resolutions (22 and 53 regions) and a global transport model are employed in this study. A realistic satellite measurement frequency distribution, obstruction of observation due to clouds, an observation error model for natural CO2 variability have been used. The satellite measurements are valuable to constrain the source uncertainties for the tropical lands with no existing observations. The optimal extension of the surface network appears to be the more effective way to reduce average inverse model uncertainty under both low and high inverse model resolutions; however, in high-resolution inversion the relative merit of the satellite data is higher than that in the low-resolution case. Our study suggests that the CO2 observation at greater data density can constrain the inverse model fluxes better. We show that the systematic errors in satellite observations can lead to significant shifts in the inverse model-estimated fluxes.