The increasing world demand for cashew (Anacardium occidentale L.) nuts and by-products generates rapid expansion of cashew cultivation across West-African countries especially in Cote d’Ivoire. This has created wealth for many smallholders. This is not to mention the pressure on forest-savanna transition zone. The aim of this study is to assess the impact of cashew production on carbon stocks. Vegetation inventory and soil sampling (0-20cm and 20-40cm) were done to estimate the above and below ground as well as soil carbon for savanna, forest and cashew plantain at different growing stages. The total carbon stocks in Mg C ha-1 were low in cashew plantations, where mature stands had 21.826 ± 3.23 (Mean ± SE), young 25.927 ± 6.53 and juvenile 16.732 ± 2.96 compared with natural vegetation (forest/woodland 64.375 ± 12.43, tree savannas 23.94 ± 3.3 and tree/shrub savannas 21.012 ± 10.12). There was no significant difference in soil organic carbon and total soil carbon stocks under different land use types, except between forest (24.67 ± 5.37 Mg C ha-1) and tree/shrub savanna (8.92 ± 1.57 Mg C ha-1). This implies that cashew expansion is of higher threat to more woody vegetation which has serious implication in terms of conservation and carbon sequestration. There is therefore a need for a more sustainable management approach to cashew agriculture practices to ensure optimum production for farmers, while conserving the forest-savanna ecosystem.

For the sustainable use of groundwater, this study analyzes groundwater potential in Western Cameroon Highlands using artificial neural network model (ANN), GIS tools and remote sensing. Twelve factors believed to influence the groundwater occurrence were selected from literature and field investigations and used as input data. Satellite ALOS PALSAR, LANDSAT OLI, SRTM data processing techniques and GIS spatial analysis tools were used to prepare these maps. Pumping rates from 189 wells were considered as groundwater potential data and randomly divided into a training and a test sets. An ANN based on the relationship between groundwater productivity data and the above factors was implement on MATLAB. Each factor