Soil erosion remains the principal cause of soil degradationworldwide, soil loss continues to be a critical concern for the sustainable management of agricultural resources, and off-site sedimentation from upland sources severely affectswater qualityand ecology.Areas of intense, localized erosion, such as rills and gullies, often are the primary cause of soil loss and the dominant source of sediment yieldfrom these landscapes. Yet there remains a serious disconnectbetween the geomorphic and hydraulic drivers of soil erosion on hillslopes and agricultural fields(rills and gullies), and current field-scale prediction and management technology. This is because most modelsdo not address rill and gully erosion explicitly. This project proposes to combine laboratory, analytic, and field investigation of the emergence, evolution, and resiliency of rill networks on soil-mantled landscapes at the field-plot scale to address this disconnect. It is hypothesizedthat rill networks display rapid pattern-emergent characteristics and long-term resiliency on hillslopes, and that the location, organization, and dimensions of these rills can be predicted before any erosion takes place. This project will conduct full-scale soil erosion experiments where terrainanalysis will be combined with state-of-the-art physicalmodels and analytical formulations for soil erosion processes. The theoretical framework and terrain analysis tools developed in the experimental landscape then will applied to a field experiment, where erosion plots will be created to replicate the in-house experiments and monitored over time. The ultimate goal of this research is to develop more physically-based soil erosion prediction models that fully exploit the latest digital technologies available, and to mitigate soil and landscape degradation before it actually occurs.