This dataset was generated with a novel process-based stochastic modelling approach to investigate the productivity and sustainability of groundwater abstractions in the Precambrian basement aquifer in Ghana. The statistical distribution of the generated synthetic yield data was found in very good agreement with observed yield data from the same Ghanaian aquifer. The dataset includes more than 40,000 simulated values of maximum allowable yield and corresponding transmissivity values for different realisations of aquifer heterogeneity, net recharge values, and borehole depth. Further details about the dataset and the method of generation and collection can be found in the article by Bianchi et al. (2020) "Investigating the productivity and sustainability of weathered basement aquifers in tropical Africa using numerical simulation and global sensitivity analysis" published in the Water Resources Research journal. This research was supported by the UKRI British Geological Survey NC-ODA grant NE/R000069/1 and NE/M008827/1.