Mapping Congo Basin vegetation types from 300 m and 1 km multi-sensor time series for carbon stocks and forest areas estimation

[摘要] This study aims to contribute to the understanding of the Congo Basinforests by delivering a detailed map of vegetation types with an improvedspatial discrimination and coherence for the whole Congo Basin region. Atotal of 20 land cover classes were described with the standardized LandCover Classification System (LCCS) developed by the FAO. Based on asemi-automatic processing chain, the Congo Basin vegetation types map wasproduced by combining 19 months of observations from the Envisat MERIS fullresolution products (300 m) and 8 yr of daily SPOT VEGETATION (VGT)reflectances (1 km). Four zones (north, south and two central) weredelineated and processed separately according to their seasonal and cloudcover specificities. The discrimination between different vegetation types(e.g. forest and savannas) was significantly improved thanks to the MERISsharp spatial resolution. A better discrimination was achieved in cloudyareas by taking advantage of the temporal consistency of the SPOT VGTobservations. This resulted in a precise delineation of the spatial extentof the rural complex in the countries situated along the Atlantic coast.Based on this new map, more accurate estimates of the surface areas offorest types were produced for each country of the Congo Basin. Carbonstocks of the Basin were evaluated to a total of 49 360 million metric tons.The regional scale of the map was an opportunity to investigate what couldbe an appropriate tree cover threshold for a forest class definition in theCongo Basin countries. A 30% tree cover threshold was suggested.Furthermore, the phenology of the different vegetation types was illustratedsystematically with EVI temporal profiles. This Congo Basin forest types mapreached a satisfactory overall accuracy of 71.5% and even 78.9% whensome classes are aggregated. The values of the Cohen's kappa coefficient,respectively 0.64 and 0.76 indicates a result significantly better thanrandom.

[学科分类] 地球化学与岩石