The loss of agricultural chemicals in runoff from agricultural land is a major cause of poor surface water quality in the United State. A technique utilizing climatic, hydrologic and soil survey information was developed to estimate the impact of agricultural watersheds on natural water resources. The objectives of this study were to apply this technique on the Wagon Train (WT) watershed to predict loss of eight elements (Al, Fe, Si, Cd, Cu, Ni, Pb, and Zn) by runoff from soils, and to estimate elements loading into WT reservoir. The predicted losses of Al, Fe, and Si by runoff were 25.3, 13.7, and 28.9 kg/ha/yr, respectively. The corresponding values for Cd, Cu, Ni, Pb, and Zn were much smaller at 0.61, 52.0, 21.3, 1.40, and 37.4 g/ha/yr, respectively. These data give a total annual loss (from the entire watershed) of 98.3, 53.2, and 112 Mg for Al, Fe, and Si, respectively. The total annual loss was 2.4, 202, 82.7, 5.4, and 147 kg for Cd, Cu, Ni, Pb, and Zn, respectively. The predicted Cd, Cu, and Pb concentrations in runoff were in reasonable agreement with the concentrations observed in the main stream in the watershed. However, the predicted concentration for other elements (Al, Fe, Si, Ni, and Zn) investigated in runoff was greater than that observed in the stream water. Elements uptake by algae, weeds, and aquatic plants and/or precipitation due to high pH in water might explain the lower element concentrations. We concluded that the technique could provide an estimation of elements loss in runoff from agricultural watersheds. The loading into surface water bodies could be predicted for Cd, Cu, and Pb. For other elements (Al, Fe, Si, Ni, and Zn), the loading could be estimated when factors affecting elements concentration in streams are considered.