As a result of past erosion problems on the Zuni Indian Reservation in western New Mexico, the US Congress in 1990 authorized the Zuni Tribe to begin a program for watershed rehabilitation. This paper describes an approach to rank the most appropriate watersheds for rehabilitation for the Zuni Reservation. The approach was based on data collected during a 3-year study on geomorphic and anthropogenic characteristics of the Rio Nutria Watershed, including data on (i) arroyo cross-sectional changes, (ii) erosion-control structures, and (iii) sheetwash erosion. Results of this 3-year study indicated that 61 of 85 channel cross-sections aggraded and channels with lower width-to-depth ratios eroded. Results on assessment of erosion-control structures, some dating back to the 1930's, indicated that 60% of earthen dams and 22% of rock-and-brush structures were breached or flanked in the Rio Nutria Watershed. Sheetwash erosion measured on five land-cover sites (sagebrush, pasture, chained pin??on and juniper, unchained pin??on and juniper, and ponderosa pine) indicated chained pin??on and juniper sites and pasture sites had the highest volume-weighted sediment concentrations of 13,000 and 9970 ppm, respectively. Based on interpretations of the 3-year study in the Rio Nutria Watershed, a two-stage approach was developed to rank the most appropriate watersheds for rehabilitation on the Zuni Reservation. In the first stage, the reservation was divided into eight major watersheds, which were ranked according to the most potential for erosion. In the second stage, the watershed with the most potential for erosion was divided into sub-basins, which were ranked according to the most potential for erosion. Quantitative and qualitative information on physical and anthropogenic factors were used at each stage to rank the watersheds. Quantitative physical data included headcut density, percentage of bare ground, percentage of chained area, channel width-to-depth ratio, change in channel density from 1934 to 1988, and sheetwash erosion rates. Qualitative physical data included erosion rankings on the main channels, tributaries, and entire basins. Anthropogenic data included density of dirt roads and condition of erosion-control structures. A community survey and agricultural acreage were also used in the selection process. The first stage analysis resulted in the selection of the Rio Nutria Watershed as the most appropriate major watershed for rehabilitation. In the second stage, the Rio Nutria Watershed was divided into 15 sub-basins; the analysis indicated the highest priority sub-basins for rehabilitation were Benny Draw, Coal Mine Canyon Draw, and Garcia Draw.