Much of the western United States is considered to be rangeland. Rangeland areas produce a diverse mix of benefits and products, and their overall health, in an ecosystem context, is of national importance.Because sediment yield from a watershed is an integrated expression of all soil erosion and sedimentation processes occurring within it, it is logical that we seek to quantify and interpret sediment yield in the context of soil/site stability and watershed function as measures of rangeland health. Depth integrated suspended sediment samples were combined with runoff measured using flumes to calculate sediment discharge and yield from two experimental watersheds in the southwestern USA. Sediment yield estimates for individual runoff events were summed to produce estimates of annual sediment yield from these two rangeland watersheds. Estimated annual sediment yield data were then combined with the concepts of sediment delivery ratio and soil loss tolerance to assess soil/site stability at the watershed scale. Analyses suggest that using sediment yield estimates from distributed watershed processes with time-space averaged soil loss tolerance values is inconsistent. Thus, new distributed soil/site stability criteria are needed to replace the soil loss tolerance concept in assessing the health of rangeland watersheds. Sediment transport/yield models are used at interior points in a watershed to simulate distributed sedimentation processes. However, application of these models requires calibration and validation data and is thus dependent upon the availability of sediment concentration and yield databases. Therefore, additional efforts are required to build sediment yield databases through rescue of historical data along with continued measurement and monitoring at existing and new sampling sites. Much of the western United States is considered to be rangeland. Rangeland areas produce a diverse mix of benefits and products, and their overall health, in an ecosystem context, is of national importance.Because sediment yield from a watershed is an integrated expression of all soil erosion and sedimentation processes occurring within it, it is logical that we seek to quantify and interpret sediment yield in the context of soil / site stability and watershed function as measures of rangeland health. Depth integrated suspended sediment processes were combined with runoff measured using flumes to calculate sediment discharge and yield from two experimental watersheds in the southwestern USA. Sediment yield estimates for individual runoff events were summed to produce estimates of annual sediment yield from these two rangeland watersheds. delivery ratio and soil loss tolerance to assess soil / site stability at the watershed scale. Analyses suggest that using sediment yield estimates from distributed watershed processes with time-space averaged soil loss tolerance values ​​is inconsistent. Thus, new distributed soil / site stability criteria are needed to replace the soil loss tolerance concept in assessing the health of rangeland watersheds. However, application of these models requires calibration and validation data and is therefore dependent upon the availability of sediment concentration and yield databases. Therefore, additional efforts are required to build sediment yield databases through rescue of historical data along with continued measurement and monitoring at existing and new sampling sites.