METHODS



Sites from the reports of the CARP and SARG surveys conducted in the early 1970s by Fred Plog and colleagues were compiled and entered in an ArcView 3.2 GIS table format. Their position on a map of the Apache-Sitgreaves National Forest was then plotted, and these data were compared to those available from the US Forest Service (as available through the AZSite directory). Sites that did not appear to be duplications in the latter database were also included in the study. A total of 953 sites were eventually used to generate the maps utilized in this analysis.

Based on the available data from the site reports, site size was also recorded in an effort to detect any patterns in site distribution related to their size. As well, each site was assigned to a specific 'Era', on the basis of the recovered artifactual material they contained. These eras were broadly defined so as to encompass the variability inherent to each. They are:

  1. Non-Ceramic: Potentially Archaic in age (< 200 AD);
  2. Early Ceramic: Early Anasazi/Mogollon, Pithouse (200 - 700 AD); and
  3. Late Ceramic: Pueblo Period (700 - 1200 AD).

I also generated other categories included under the heading 'Era,' to help in the analysis and palliate certain deficiencies in the data:
  1. Historic: Associated with modern artifacts (tin cans, etc.);
  2. Checkdams: When checkdams were observed, but no other archaeological evidence was reported;
  3. Ceramic: When ceramic was recovered, but not described in enough detail to permit a specific temporal assignment; and
  4. Unknown: When no information was given as to the artifactual material associated with a reported site, thus preventing a temporal designation.

The vast majority of sites unfortunately had to be labeled as 'unknown,' due to the very uneven quality of data recording in the CARP and SARG site reports. For analytical purposes, 'unknown' sites were disregarded, as were historic ones, since neither category could be used to provide a better understanding of the prehistory of the area.

Using ArcView 3.2, a number of relevant variables (i.e., 'themes) were studied in relation to one another and to site distribution to see if any significant patterns of association emerged.The themes were geology, vegetation cover, soil composition and watersheds. I also included the major rivers in the area, all of which were buffered with 0.5, 1.0 and 3.0 km zones to give an at-a-glance view of how distant sites of different periods were to major water sources.

IntroductionProblemsAnalysis & InterpretationAbout me!