Global atmospheric carbon monoxide budget 2000–2017

We use a multi-species atmospheric Bayesian inversion approach to estimate the global carbon monoxide (CO) budget during 2000-2017, constrained by satellite retrievals of the total column mole fraction of CO, formaldehyde (HCHO), and methane (CH4). Here we provide monthly global gridded maps of surface CO emissions, CO chemical production, and CO chemical sinks. Data from three inversions are generated.1. Inversion #1 (2000–2017) is constrained by Measurements Of Pollution In The Troposphere instrument (MOPITT) v7 CO total column;2. Inversion #2 (2005–2017) is constrained by both MOPITT v7 CO column and Ozone Monitoring Instrument (OMI) v3 HCHO column; 3. Inversion #3 (2010–2017) is constrained by MOPITT v7 CO column, OMI v3 HCHO column, and Greenhouse gases Observing SATellite (GOSAT) column-averaged dry air mole fractions of CH4 (XCH4). Based on a comprehensive evaluation, we think Inversion #3 has a more realistic representation of the global CO budget than Inversion #2 does, and Inversion #2 is better than Inversion #1. It is appropriate to use Inversion #3 and Inversion #2 for a long-term trend analysis but they are both limited to short periods. If Inversion #1 has to be used due to its long-term temporal coverage, caution needs to be taken that the decreasing trends of anthropogenic CO emissions are probably underestimated and the growing CO chemical production is not well represented over anthropogenic source regions. For more details, please see our paper entitled "Global atmospheric carbon monoxide budget 2000–2017 inferred from multi-species atmospheric inversions".