Despite of Lesotho having abundant water, it is still faced with freshwater challenges as the current water supply cannot meet the needs of the population. In addition, water has become the main driver of development in Lesotho as it is one of the few natural resources of economic importance hence construction of water storage dams has been identified as a development strategy. Modelling of water flow is one of the techniques used in describing the movement of water and determining flow accumulation within the catchment. Rainfall-runoff modelling in Lesotho has been based on traditional methods which only focus on the discharge at the outlet neglecting the distribution of runoff over the catchment. GIS enables modelling of spatial variability hence this study is aimed at determining flow distribution and accumulation within Phuthiatsane Catchment and estimating runoff potential in a GIS environment. In order to achieve this, ArcHydro extension of ArcGIS was used in the determination of flow distribution and subsequent catchment delineation while ArcCN-Runoff tool was used to determine the potential runoff based on land cover, soil type and amount of rainfall. The delineated catchment covers an area of 468 km2 and has an average runoff of 30.943 MCM. Suitable dam sites, volume and the areas that would be submerged were then identified using other ArcGIS tools. It was concluded that GIS can produce accurate hydrological modelling results for Phuthiatsane Catchment. Simulation of dam sites and storage capacities has also proven to be efficient in GIS environment and a 70m high dam with a storage capacity of 327.92 MCM was considered to be the most suitable. It is recommended that further research should include ground truth surveys for the validation of results. Further research should also incorporate stakeholder concerns in deciding on the location and size of a dam. / Thesis (M.Sc.)-University of KwaZulu-Natal, Westville, 2010.