Assessment of regional carbon monoxide simulations over Africa and insights into source attribution and regional transport

This study assesses the quality of carbon monoxide (CO) simulations from the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) and analyzes different processes controlling CO distribution in Africa. Sixteen CO tracers are implemented in WRF-Chem to track CO from different sources and regions. The model captures the seasonal cycles of surface temperature fairly well (r > 0.75 at most sites) and reproduces key features and variability of precipitation, wind speed, and planetary boundary layer height with errors typically seen in application of a regional model over such a large domain. The model also captures CO concentration hotspots and vertical gradients but underestimates the Measurements of Pollution in the Troposphere (MOPITT) retrieved total column with a larger underestimation in northern hemispheric Africa. The model also underestimates MOPITT and aircraft observed CO profiles throughout the troposphere. We discuss outdated anthropogenic emissions, low biomass burning emissions, and well-known low biases in global models of CO as potential contributors to the discrepancies between the regional model and MOPITT CO. Source contribution analysis of surface CO showed that all sources contribute to CO in Africa with anthropogenic emissions dominating in the urban regions, fire emissions dominating in central Africa, CO inflow from domain boundaries dominating in North Africa and the oceanic regions, and photochemical production dominance aligning with seasonal changes in sunlight. We also show that regional transport of CO emissions between African regions also play a very important role in controlling surface CO distribution in Africa. In some regions, more than 50% of the anthropogenic or biomass burning emitted CO comes from sources located outside those regions, highlighting the need for regional coordination and cooperation for air quality management in Africa.