Severe sediment-laden problem has been the hot issue in Korea. It was assumed that agricultural activities and landslides were the primary causes of these problems in watersheds. The USLE-based systems have been widely used in soil erosion studies. However the GIS-based USLE modeling system has limitation in estimating USLE L factors at a watershed scale. In this study, the SATEEC L module was developed to reflect the slope length segmentations in the fields. The SATEEC L module was applied to the study watershed to analyze the effects of using the SATEEC L module on estimated sediment. As shown in the comparisons between SATEEC estimated sediment with SWAT values, the SATEEC GA-SDR module derives the SDR with reasonably acceptable accuracies. However, it is worthy to note that the soil erosion using the SATEEC L module for the study watershed was lower than that without using the SATEEC L module by 25%, although the SATEEC estimated sediment values with and without using L module match the SWAT sediment values with similar accuracies (over/under estimation). This is because the SATEEC GA-SDR module estimates lower SDR in case of greater soil erosion estimation without the L module and greater SDR in case of lower soil erosion estimation with the L module. This indicates that the SATEEC input parameters, especially L factor, need to be prepared with care for accurate estimation of SDR at a watershed scale and for accurate evaluation of BMPs in the watershed.