Sand Canyon Pueblo

True north (14 degrees declination) (USGS Quadrangle Map: Woods Canyon, Colorado, 7.5 minute, 1965 [photoinspected 1979]).

Grid aligned on true north.

The remains visible on modern ground surface were mapped in 1983 with an alidade and plane table. During the years of excavation, a transit was used to establish a grid, to set in excavation units, and to establish vertical datums. Several datums were established; a primary datum was set at 1000N 1000E, 100 m elevation. A photogrammetric survey of Sand Canyon Pueblo was completed in 1985, and the resulting map served as the basis for the final topographic map drafted in AutoCAD.

In accordance with Bureau of Land Management (BLM) stipulations, vegetation was cleared only in the areas that were excavated. Live trees were not cut or otherwise modified unless necessary. When tree roots were partly exposed, attempts were made to preserve as many as possible. Also in accordance with BLM directives, some of the vegetation (brush and tree limbs) cleared from excavated areas was deposited in an arroyo between Kiva Suite 501 and the great kiva.

In accordance with Bureau of Land Management (BLM) stipulations and Crow Canyon policy, all excavated areas were backfilled after the completion of recording. Shovels and wheelbarrows were used for most backfilling, although a front-end loader was used to backfill parts of Kiva Suites 102, 306, and 1004. The lowest level of each excavated unit (except some units that ended on bedrock) was lined with strips of landscape fabric before earth and rocks were used to fill the area. In addition, empty aluminum soft-drink cans, usually containing a contemporary penny, were placed on structure floors, in hearths, and occasionally in other features. Loose building blocks often were stacked against standing walls before the units were backfilled. The materials removed during excavation were used for backfilling, although no attempt was made to return specific deposits to their exact place of origin. Attempts were made to return the ground surface to its original contour and general appearance. If the area had been covered with stone rubble before excavation, stone rubble was used to cap the backfilled area afterwards. Care was taken to return stone with lichen-covered faces on top, with the lichen surfaces up. Some areas were reseeded according to BLM standards, using specified seed mixes. Other areas were rapidly revegetated as the result of natural plant growth. With the exception of the return of trees, the areas of the site that were excavated and backfilled returned to a "preexcavation look" within two to three years.

Kivas were usually identified on the basis of circular depressions in areas of stone rubble. Towers were identified by circular or D-shaped piles of rubble. Most roomblocks were indicated by concentrations of rubble, although it usually was not possible for us to discern the outlines of individual rooms on the modern ground surface. Some refuse was visible on the talus slopes.

As part of his axe study, Peter Mills established three transects from which he collected all artifacts found on the modern ground surface (Mills 1993*1; also see Database Map 4315 for transect locations). Otherwise, no artifacts were collected from the modern ground surface in areas of the site that were not excavated.

An attempt was made to excavate areas of extensive disturbance separately from the surrounding undisturbed areas to avoid mixing materials from interpretable and uninterpretable contexts. Deposits that were displaced naturally--for example, walls that collapsed as the result of freezing and thawing during the winter--were excavated and recorded separately from in-place deposits. Areas of minor animal or other natural disturbance generally were not excavated separately from the surrounding areas, although if the displacement of materials was severe enough to limit interpretation, the fill was coded as "mixed."

Crow Canyon staff and participants established and maintained a field camp, conducted archaeological excavations (which resulted in the deposition of "backdirt" and, later, the refilling of excavated areas with backdirt and rocks), developed a trail system for excavators and visitors, constructed a stone wall as part of a study of prehistoric labor requirements, and cut down a tree with a stone axe as part of a study of artifact use (Mills 1993*1). The field camp consisted of two mobile homes, an attached kitchen and eating area, parking areas, trails, tent platforms, a water tank, a water line, a campfire area, a temporary "sweat lodge," waste-water pits, and portable chemical toilets. All facilities, as well as the experimental stone wall, were removed at the completion of excavations, and all areas disturbed by our work were backfilled, contoured, and reseeded. The trail in and around the site was left to facilitate future visitation and interpretation.

During mapping in 1983, a number of areas of historic disturbance were noted. Pothunters' holes were evident in a structure in Architectural Block 1400, in the upper portion of the fill of Kiva 1206, in a large section of Midden 103, in most of Room 1505, in the southern end of the plaza, and in the midden downslope from Architectural Blocks 300 and 400. Overall, these disturbed areas make up less than 1 percent of the surface area of the site. In the early 1960s, a temporary access road was constructed to allow for seismic testing and the digging of a stock pond; the construction of this road disturbed an unknown number of structures in Architectural Block 1100. Local residents report that the spring was dynamited in the early 1900s by a nearby homesteader. The Bureau of Land Management modified the spring in 1985 by inserting a barrel filled with crushed stone as a collection receptacle and running a hose downhill to a spigot. The barrel filled with sediment and was rendered useless in September 1985. A small, conical "sweat lodge," located just outside the site-enclosing wall near Kiva 107, is purported to have been built in 1969 by students from the University of Colorado.

Architectural stones, such as building rocks and door slabs, were not collected. Petroglyphs in walls were considered features and were not collected.

Constructed surfaces, use-compacted surfaces, collapsed roof surfaces, other use surfaces (e.g., ephemeral surfaces).

Artifacts were provenienced with a surface if they were in direct contact with, or immediately above, a surface. In some cases, artifacts found farther above a surface were interpreted to have been surface-associated, especially if they were oriented horizontally and consisted of complete tools or vessels. Artifacts resting on other surface artifacts were also interpreted to be surface-associated.

All burned and unburned wood with enough (15 or more) annual rings to be potentially datable was collected, except when we excavated burned roof deposits. In the latter case, when many pieces of wood were obviously from the same beam, usually only one sample was collected (occasionally two); the remaining pieces were mapped, and some were collected as vegetal specimens. Tree-ring samples were collected as soon as possible after the wood was exposed, and each sample was securely wrapped with cotton string.

Samples were collected from hearths and other burned areas only if, on the basis of intensity of burning, clay content, and degree of disturbance, the sediment was believed to be potentially dateable.

Sediment samples for flotation were collected from areas where plant remains were expected to occur (for example, hearths). These samples were usually 1 liter in volume, but exceptions were made when less than 1 liter of sediment was available, or when so much sediment was available that it was possible to collect multiple 1-liter samples. We consistently sampled ash deposits in thermal features and in secondary-refuse deposits. Samples were also collected from burned roof contexts, the contents of whole or nearly whole vessels, and from use surfaces where there were sealed, floor-associated deposits. We avoided collecting samples from mixed or disturbed deposits.

Sediment samples collected for the purpose of recovering fossil pollen were taken from undisturbed contexts when it was expected that the results of analysis might contribute to our understanding of a particular depositional event or sequence. Samples from use surfaces were restricted to deposits that were sealed under floor artifacts. Sample columns were collected from selected postabandonment sediments that had been deposited naturally; it was hoped that these samples would allow us to identify the succession of plants in the site area after abandonment. In addition, pollen samples were taken from places (such as niche surfaces) where one might expect enhanced cultural deposits of pollen.

Ash samples were collected from hearths in several kivas as part of a study to identify different kinds of fuelwood (Pierce et al. 1998*1).