These data were gathered for the purpose of research in Work Package 4.5 within the context of the BiomassWeb Project (https://biomassweb.org/). The data are on agricultural land use, resource management practices and production in small-scale maize systems in Bolgatanga and Bongo Districts, Upper East Region, Ghana. The data were collected by Francis Molua Mwambo, using semi-structured questionnaire to interview maize farmers in 2015. The data are based on the interviewees' estimate, which were collected as feedback to the questionnaire during a field survey in the study area.

The data was collected in three districts (Kersa, Omonada and Bako-Tibe) in Oromia regional state, Southwestern Ethiopia. The data was collected from 228 farmers during October-November 2015. The questionaire was developed after a series of focus group discussions and key informant interviews in the study area.

Land degradation in this dataset is defined as the persistent reduction or loss of land ecosystem services, notably the primary production service. The long-term trend of inter-annual mean Normalized Difference Vegetation Index (NDVI), over the period 1982–2006, is used as a proxy for the persistent decline or improvement in the Net Primary Productivity (NPP) of land, thereby delineating land degradation hotspots. The Global Inventory Modelling and Mapping Studies (GIMMS) dataset of 64 km2-resolution of NDVI data employed has been corrected for rainfall variation effect and atmospheric fertilization effect in addition to the masking of ineligible pixels. The data is used to calculate the area of NDVI decline in km2 and in percentages for corresponding land covers. The dataset provided contains different files (GIS files, a table and a report). The report,"BIOMASS PRODUCTIVITY-BASED MAPPING OF GLOBAL LAND DEGRADATION HOTSPOTS", contains a detailed description of how the global land degradation hotspots, based on biomass productivity, were identified. A summary of the “ground truthing” methods is also presented. The GIS files map the global land degradation hotspots versus main land cover/use types. In addition, the GIS files map the areas with above ground biomass improvement. The area (km2) of long-term (1982-2006) NDVI decline (with correction of RF and AF effects and masking of saturated NDVI zone) versus main land cover/use types counted for each country are then presented in the excel table.