Runoff and soil erosion are two important factors in water resources management. These factors should be considered while planning for watershed management. But, researchers face difficulties in the modeling of runoff and soil erosion in the watershed due to their nonlinearity and scaling effect. The hydrological connectivity may decrease due to infiltration and deposition of sediment in watershed. Hence, it is important to use tools like rainfall-runoff simulation model to estimate runoff and soil erosion from the small watershed. There are a lots of models are available to predict runoff and sediment from the small watershed. In this study, Natural Resources Conservation Service Curve Number (NRCS-CN) method is used to predict the depth of runoff and peak discharge from an ungauged small watershed by coupling remote sensing and GIS software with HEC-HMS. By using the peak discharge and depth of runoff the sediment yield has been estimated by using Modified Universal Soil Loss Equation (MUSLE). The regression analysis method has been used to derive the relationship between the depths of rainfall sediment yield from the small watershed. The frequency analysis has been performed to predict the depth of runoff and sediment yield in the different return period. © IAEME Publication.