A regional-scale watershed assessment using the relative risk model has now been performed on each of the watersheds that are part of the Long-Term Receiving Waters Study (LTRWS) to generate patterns of risk at each study site. The watersheds assessed included the lower McKenzie and mid-Willamette rivers in Oregon, USA; Codorus Creek in Pennsylvania, USA; and the Leaf River in Mississippi, USA. In each case, the goal was to put the LTRWS into a watershed and regional context, including multiple sources, stressors, habitats, and assessment endpoints. In each instance, the relative risk model provided measures of relative risk and testable hypotheses about patterns within the watersheds. There were similarities among the 3 risk studies. First, land use within the watershed was the most important driver of risk in each instance, even compared with point sources. Second, the list of endpoints was similar for each risk assessment. Water quality was held as important, along with populations critical to recreation or commercial use. At each site, data from the LTRWS and other sources could be used to support the risk patterns predicted from the risk assessment process. However, the sampling design of the LTRWS prevented the confirmation of specific cause–effect relationships. A fundamental conclusion is that risk assessment, using an adaptation of the relative risk model, should be performed as an integral part of any watershed assessment and management effort. These initial watershed risk assessments have led to similar assessments across the world.