At an altitude of nearly 5000 masl in sandy soil, this area is far above the growing zone even for tubers. If these were residential terraces, where were the ceramics? One possible explanation was provided from three¹⁴C samples from the workshop test unit [Q02-2u3] that revealed that the workshop occupation belonged to the preceramic and very early ceramic period. It appears that that the dominant component of this site is from prior to the use of ceramics in the area.

The terrace margins are generally highly eroded and ill-defined in places, and the focus in 2003 was therefore on mapping terraced zones as several large polygon areas rather than attempting to map each terrace as a linear feature. The upper terraced area [A03-334] was particularly eroded, but faint traces of intermittent terraces were apparent. The terraces walls, and wall bases that appear to have been small circular structures, are single walled with no mortar. The sole exception was a corner of doubled-walled construction made of ground fieldstone in the lower terraced area [A03-276] where the corner of a structure of cut-stone masonry of possible Late Horizon date was located, a feature that is described later in the Late Prehispanic Block 1 section.

The eroded terraces of A03-275 are generally 20-50cm in height and are constructed with fieldstones of a variety of sizes. Typically, a few large boulders will form the general structure of the terraces, and then small level surfaces would be constructed by building terrace walls of the flat local lava rock. It is difficult to date these constructions but due to the presence of sherds from several Inka plates, and a possible LH feature, these terraces are further discussed in the Late Prehispanic section titled A03-275 "Maymeja 5", along with several photos of these constructions.

In the sector of A03-126, above the bofedal margin, several small circular structures [A03-335] of possible Early Agropastoralist age were identified. These structures consist of circular wall-bases with concentrations of obsidian eroding downslope from the interior area. Two adjacent circular constructions were in the middle of an eroded terrace, but the predominant pattern was for small circular constructions measuring 2-3m in diameter to be built adjacent to rock outcrops that appear to offer protection from the western winds. These small wall-bases were observed along the expansive rock outcrop that extends just south of the bofedal. No hearths or bones were observed in this area, although bone preservation would probably be very poor in this exposed area.

Figure 6-53. Base of structure [A03-335] is formed by fifteen large, partially buried stones and measures 2.5m in diameter.

These structures are being interpreted as residential constructions or windbreaks occupied on short term basis by obsidian procurers who were allowing their animals to graze while they quarried and reduced obsidian from the Maymeja area, and perhaps dug the quarry pit Q02-2. Herders would presumably have had sufficient animal hides and woolen textiles to insulate stone walled structures from the penetrating winds.

It is also conceivable that these are bases for large circular LIP chulpas, as the wall bases are sufficiently large. This is unlikely, however, as there were no LIP ceramics in the area, and the obsidian flaking debris eroding downhill strongly suggests that obsidian reduction was occurring inside these circular structures.