Groundwater is a valued resource of limited extent both in quantity and in space. In order to ensure its sustainable use, appropriate evaluation of its availability is required. This study aimed at solving groundwater problem fraught with uncertainty by considering recharge as uncertain input parameter. The developed groundwater flow model was later calibrated using automatic calibration software PEST within MODFLOW environment. Head values from groundwater level measurement data were used as calibration targets. The calibration resulted in normalised residual mean square (NRMS) value of 1.828%, mean error (ME) of 0.999 m, mean absolute error (MAE) of 6.701 m, Root Mean Square Error (RMSE) of 8.499 m and correlation coefficient (Cor) of 0.997. The field monitoring borehole dataset of 2011 was used to validate the model. The validated model resulted in a Root Mean Square Error (RMSE) of 8.601 m and NRMS value of 1.85% which compares very well with the NRMS value of the calibrated model whose value was 1.828%. All set criteria were within acceptable limits for both calibrated and validated model. Thereafter, stochastic uncertainty analyses were performed by randomly sampling recharge realization and running a series of Monte Carlo (MC) simulations ranging from 50 to a maximum of 1000 realization of recharge fields. Stochastic model results were compared to those of deterministic model. Groundwater potential results were compared for both deterministic and stochastic scenarios. Deterministic scenario resulted in groundwater potential value of 1122.92 Mm3/year while stochastic scenario resulted in a value of 1137.61 Mm3/year, an increase of 14.7 Mm3/year. This implies that there is potential for more groundwater exploitation in the study area. Thus, the developed stochastic model can be used as a decision support system (DSS) for water resource management.