These are the largest sites in Block 2. They consist of pastoral facilities, associated residential structures and adjacent middens, and an artifact scatter extending along the margins of the grazing area. These sites are commonly found on a sloping area or a raised area along the margins of the pampa where the drainage is adequate. Animal control structures were frequent on these sites, with the majority of them taking the form of maintained or abandoned corrals of varying sizes. At the sites designated as "Pastoral Bases", corral structures were mapped and they range in area from 62 m2to 4,306 m2with a mean corral size of 671m2. Due to high rates of erosion, it was difficult to confirm from the soil consistency if it was largely dung soil inside of structures that appeared to be long-abandoned corrals. However, the raised area of these corrals are effectively low mounds that may be the result of accumulation of dung over long time periods, combined with some building up and filling in of the raised area to lift the corral areas off of the level of the pampa to improve drainage and avoid inundation during the wet season.
These structures on the puna edge appear to be pastoral facilities containing the bases of walls formed by rocks between 30 cm and 100cm in size that create a circular or elliptical enclosure. The area will often contain smaller enclosures that probably served more specifically as corrals, subdividing the protected area. These corrals might be occupied simultaneously or sequentially, and could contain individual herder's animals, or they may segregate the herd into sex and age categories as a part of pastoral management strategies (Flannery, et al. 1989;Flores Ochoa 1968). The presence of such large rocks at the base of these walls is perhaps explained by the need to keep small animals, such as young camelids, inside and the need to keep small predators out. Modern herder-built walls are often solidly constructed along the base and only along the top of the wall do smaller rocks get used.
These wall bases could be the remnants of corrals used by seasonal residents of this rich pasture region, or if they were used by passing caravans, the corrals could have been important facilities as part of a multi-day rest stop for caravans (Lecoq 1988: 185-186;Nielsen 2000: 461-462, 500-504;West 1981: 70).
The perimeters of pastoral bases often have dense artifact scatters, including pottery fragments from a variety of time periods, however there was a relatively low frequency of high density lithic loci inside of pastoral bases. When all of the high density loci are considered, only 30 (23%) of the high density lithic loci were located in sites classified as "pastoral bases".
High density lithic loci in this region primarily consist of obsidian flakes and it appears that, despite being located off of the pastoral base sites, these high lithic concentrations still seem to be associated with early pastoral occupations. This inference is based on two aspects of these high density loci: (1) most of the pre-Series 5 projectile points are not obsidian, while most of the Series 5 points are obsidian, and therefore if point production was occurring they were probably making Series 5 points; (2) if merely flakes were being produced, such production was probably linked to pastoral butchering, and therefore the obsidian is again linked to the pastoral occupation of the area. Why, then, were the high concentrations located outside of the pastoral base areas as noted above? One possibility is that one may be seeing a site maintenance pattern where some lithic production is occurring offsite during pastoral herding so that the residential base is not littered with sharp flakes.