The objective of this project was to model surface water transport of radioactivity released from the Fukushima Nuclear Power Plant. The first step in the process was to build a river network for Japan with associated flows and velocities. This was accomplished by using the Geospatial Stream Flow Model (GeoSFM) in conjunction with digital terrain data from the Shuttle Radar Topography Mission (SRTM), land use and soil data from the US Geological Survey and Food and Agricultural Organization, respectively, and rainfall and potential evapotranspiration (PET) data from the National Ocean and Atmospheric Administration. Benchmarking of the hydrology (flow) model was done using data from the Global Runoff Data Centre (GRDC). The GeoSFM model was run with precipitation data (CMORPH daily rainfall) and PET for the year 2006. The results were compared with the observed data downloaded from the GRDC site. Average flow for the year was used for the comparison between the observed and simulated flows. The results were in good agreement considering the fact that the observed data was averaged for more than one year. The Incident Command Tool for Drinking Water Protection (ICWater) model was used for the radionuclide transport. The hydrologic inputs for ICWater were provided by GeoSFM. For radionuclide transport validation, observed data in the river reaches is needed. The observed radionuclide data is available for air, dust, soils, leafy vegetables, and at a few places for drinking water in a 50 km radius around the plant. An Internet search was performed for surface water contamination around the plant but none was available. Because of the lack of surface water quality monitoring data, a two step approach was used for radionuclide transport modeling to determine the following parameters: (1) time of travel along major river reaches in the 50 km radius zone and (2) dilution of contaminants in the 50 mile radius zone. For travel time, a continuous release of one hour duration was assumed. The concentration of release was assumed as 100 mg/l and the flow associated with the release was assumed to be 0.1 cfs. Under these conditions, travel times ranged between 5 and 22 hours and dilution factors ranged between 3 and 14, depending on the river being modeled. Two additional test cases were performed for Cesium-137 deposition in two watersheds north of the plant. This simulation was performed using an actual rain event on 4-24-2011. The rain file contains a global estimate of 24-hour accumulated precipitation in GIS GRID format with a spatial resolution is 0.25 degrees. The ICWater model was run for 24 hours and the downstream trace from the simulation and breakthrough curves were calculated for two rivers that discharge to the ocean. Export files were created to provide a source term to the SHARC coastal transport model.