Soil erosion is one of the most serious environmental problems commonly in over the world, which is caused by both natural and human factors. GIS and remote sensing technique has recently become a powerful tool for investigations of natural resources and environmental situation. This research focuses on estimation of global soil erosion by the RUSLE model with the use of a comprehensive global dataset. The accuracy of the estimate mostly depends on the available information related to the study area. Presently available global datasets were used in this study. As the desired objective of estimating soil erosion by water at global scale, the application of RUSLE has shown its positive applicability on large-scale estimates. Global datasets of land cover, digital elevation, soil property and precipitation act as indispensable database in getting erosive parameters used in the RUSLE model. Data sources and data processing methods are carefully discussed in this paper. The study has provides a global estimate of water soil erosion potential with 0.5-degree grid resolution. Regional validations and evaluations have been carried out by different ways. The global mean of annual soil erosion by water was estimated as 1150 ton/ km², which agrees with several results obtained in different regions.