Tracing the origin and ages of interlaced atmospheric pollution events over the tropical Atlantic Ocean with in situ measurements, satellites, trajectories, emission inventories, and global models

During a west to east crossing of the tropical Atlantic Ocean in October-November 2002 on R/V Meteor (M55), carbon monoxide (CO) and ozone were continuously monitored, and pressurized air samples were collected and later analyzed in the laboratory for various volatile organic compounds. A sequence of alternating CO and propane rich events were observed over the east Atlantic, the events of enhanced carbon monoxide being out of phase with those observed for propane. A combined study of air mass origin (back trajectories and backward emission sensitivity calculations) and source region distribution comparison (CO satellite data from MOPITT and propane emission data from the EDGAR database) showed that the CO events were due to African biomass burning emissions, whereas the propane events were due to industrial emissions from areas of northern Africa. Both events were associated with elevated ozone. A comparison of the measured concentrations of CO and propane with those simulated by the global Model of Atmospheric Transport and Chemistry-Max Planck Institute for Chemistry (MATCHMPIC) shows that the model reproduces the general longitudinal gradient observed for both compounds and simulates elevated CO concentrations during the pollution events. However, it systematically overestimates the CO mixing ratios. It is suggested that the northern African biomass burning emissions used in the model are not distributed correctly (incorrect timing) and, in particular, that too high emissions from the region "northern Sudan-Sahel" are used for this period. The model does not capture the influence from industrial emissions from northern Africa, which may be caused by too strong diffusion of the plume.