Desertification in the Sahel: Towards better accounting for ecosystem dynamics in the interpretation of remote sensing images

To date, the interpretation of remote sensing images has not revealed wide-spread degradation of the vegetation in the Sahel. However, the interpretation of spectral information depends on a range of assumptions regarding the dynamics of the Sahelian vegetation as a function of rainfall variability and human management. Recent papers have presented diverging views on the vegetation dynamics of the Sahel and how these can be analysed with remote sensing images. We present a further analysis of the vegetation dynamics of semi-arid rangelands, in particular the Sahel, and the subsequent implications for the interpretation of remote sensing images. Specifically, the ecological processes driving the response of the Sahelian vegetation to rainfall variation are re-examined, and a regression analysis of NPP versus rainfall data is carried out. It is shown that the relation between the interannual variation in NPP and rainfall in the Sahel is non-linear and that this relation differs between sites with different average annual rainfall. It has been common practise in remote sensing studies for the Sahel to aggregate data from various Sahelian sites in order to obtain an average relation between rainfall, NPP and Rain Use Efficiency, and to assume these relations to be linear. This paper shows that this approach may lead to a bias in the interpretation of remote sensing images and that further work is required to clarify if wide-spread ecosystem degradation has occurred in the Sahel.