The Benguela upwelling system is one of the four large upwelling areas of the world ocean. The trade winds drive a more or less permanent upwelling fueling high primary production. This goes along with the formation of an oxygen minimum zone structuring the ecosystem and its very specific food web. Building ecosystem models of this area is a challenging task, since the oxygen conditions on the Namibian shelf depend not only from local processes but are remotely influenced by the equatorial ocean and the Angola Gyre. A coastally trapped pole-ward undercurrent serves as a link between the large oxygen minimum zone in the Angola Gyre and the Benguela system. Biological processes like microbial mineralisation of organic matter in the water column and in the sediment but also zooplankton respiration modify the oxygen conditions locally. It turns out that the proper representation of the spatial structure of the wind field is the key for realistic results for vertical and lateral advection processes. Especially the wind field structure in the area of the Kunene upwelling cell determines the pole-ward transports and thus the nutrient and oxygen supply of the Benguela. This transport is often considered as wave driven by shelf waves (Kelvin waves) of equatorial origin. Correlations between the sea level anomaly in the coastal wave guide and the sea level anomaly in the equatorial area vanish at about 15°S, but tropical water is found up to latitudes of 28°S. Hence, to understand nutrient supply and ventilation of the Benguela system locally driven pole-ward transport need consideration, either Kelvin wave induced undercurrents or wind stress curl related pole-ward flow. A realistic representation of the oxygen dynamics within the Benguela oxygen minimum zone depends on the quality of the representation of physical processes like upwelling and wave related tracer transports.
The Benguela system becomes also obvious in climate model results by the outstanding SST bias that many models produce there notoriously. Analytical and numerical models demonstrate that the key for the reduction of this bias is the aforementioned proper representation of the wind field structure and the related upwelling and pole-ward transports as well. But also a solar radiation bias and errors in the determination of the mixed layer depth contribute to the SST bias in the Benguela system. A comparitive model study evaluates the ability of several wide spread and commonly used wind products to produce realistic ocean model results. It turns out that scatterometer based wind measurements offer currently the most realistic wind products.