This research provides quantitative information on short and long-term effects of shifting agriculture on soil and spatial pattern of landscape mosaic dynamics in southern Cameroon. An analysis of this farming system led to the development of a conceptual model of the spatio-temporal dynamics of shifting agriculture, including transition matrices of rotational cycles. Geostatistical characterization of soil variability in the area showed that soil properties are highly spatially dependent even at plot level, with significant sensitivity to soil-forming factors that explained 30 to 70% of the total variation in the subsurface. Shifting agricultural land use practices accounted for 30 to 35% of the variation of topsoil. A robust quantitative multi-criteria method was developed for quantifying and selecting soil variables that are the most sensitive to these agricultural practices (in this case: pH, calcium, phosphorus, bulk density, organic carbon), considered as the minimum data set (MDS) for characterizing soil conditions in the area. Empirical models of linear/quadratic fractional rational functions were successfully fitted to time series data of these MDS variables to derive quantitative measures on temporal changes in soil with land use. Multi-spectral satellite imagery was able to map with 80% accuracy the extension front of shifting agricultural landscape and the most dynamic land cover types (crop fields, young fallows), which shift every season and every year. The research has produced a set of data and methods that can be used in combination with rare cloud-free satellite images for spatiotemporal simulation modelling of landscape dynamics in order to guide decision-making on agricultural development, land allocation for land use planning and forest resources management
