1Royal Netherlands Meteorological Institute, De Bilt, The Netherlands
2Université de Toulouse, UPS, LA (Laboratorie d'Aerologie), 31400 Toulouse, France
3Laboratorie d'Aeronomie, (UMR UPS/CNRS 5560), Observatoire Midi-Pyrénées, Toulouse, France
4School of Environmental Sciences, University of East Anglia, Norwich, UK
*now at: National Center for Atmospheric Research, Boulder, CO, USA
Abstract. We have performed simulations using a 3-D global chemistry-transport model (TM4_AMMA) to investigate the effect that continental transport of biomass burning plumes have on regional air quality over Equatorial Africa during the West African Monsoon (WAM) period in 2006. By performing a number of sensitivity studies we show that biomass burning emissions from southern Africa (0–40° S) have a strong influence on the composition of the tropical troposphere around Equatorial Africa and the outflow regions towards the west, especially between 10° S–10° N. By altering both the temporal distribution and the injection heights used for introducing the biomass burning emissions we show that changes in temporal distribution are much more important in determining the daily variability of trace gas species over the southern Atlantic than boundary layer processes. When adopting the GFEDv2 emission inventory the maximum concentrations in CO and O3 occur between 0–5° S, which coincides with the position of the southern African Easterly Jet. By comparing co-located model output with in-situ measurements made during the AMMA measurement campaign we show that the model fails to capture the tropospheric profile of CO in the burning region, as well as the "extreme" concentrations of both CO and O3 seen around 600–700 hPa above Equatorial Africa. Trajectory analysis show that the 6-hourly ECMWF meteorological fields do not allow transport of biomass burning plumes from southern Africa directly into the mid-troposphere around ~6° N. Similar trajectory simulations repeated using an updated meteorological dataset, which assimilates additional measurement data for the African region, shows markedly different origins for pollution events and reveals that the performance of the CTM is heavily constrained by the ECMWF operational analysis data which drives the model.