Irregular industrial Carbon monoxide (CO) emissions not only deteriorate global atmospheric environment but also add a large uncertainty to CO emission inventory. This work demonstrates the potentiality of the Tropospheric Monitoring Instrument (TROPOMI) data product to derive CO emissions from facility-level industrial sources. We have demonstrated CO emission estimates of 14 randomly selected facility-level industrial sources over China with uncertainties of 10%-35% from TROPOMI archive between 2017 and 2020. We found that most of the plumes reveal significantly larger emissions than the registered inventory values. The simplicity of the demonstrated plume quantification method makes it easily applicable everywhere around the globe. This study suggests that future CO satellite constellation can routinely observe point source CO emissions to support environment policy.