CRM Studies Series - Los Chozas Land Exchange Project Volume 2

PREHISTORIC COMPONENTS

Cultural and Temporal Affiliation

Thirty-six sites with Prehistoric components were recorded by survey. The temporal classification of Prehistoric components was estimated primarily from the identification of diagnostic artifacts. Forty-five percent of identified Prehistoric components date to the Late Archaic/Early Pithouse period. The remainder range from Early Pithouse (20%), to Late Pueblo (35%). Sixteen prehistoric components, composed of non-diagnostic, undifferentiated lithic material, were classified as Unknown Prehistoric. Individual site descriptions are presented in Chapter 6. Table 7.20 lists the frequency distribution of the cultural/temporal affiliations of dated prehistoric components. Table 7.20. Frequency of Cultural/Temporal Affiliations of Prehistoric Components.

Cultural Affiliation                    Temporal Range      Frequency

Late Archaic/Early Pithouse             1800 B.C.-750 A.D.       9
Early-Middle Pithouse/Late Pueblo       225/900-1350 A.D.        1
Early Pithouse/Middle Pueblo            250-1350 A.D.            1
Early Pithouse/Late Pueblo              250/900-1350 A.D.        2
Late Pithouse/Early Pueblo              850-1125 A.D.            3
Late Pithouse/Late Pueblo               850-1350 A.D.            3
Early Pueblo/Late Pueblo           	1100-1400 A.D.      	 1

Total # Dated Sites                                    		20

Archaic and Post-Archaic Assemblage Comparisons

Of the 20 prehistoric sites with dated components, nine are of Archaic affiliation and 11 are of Mogollon and Mogollon/Anasazi affiliation. Comparison of artifact frequency is expected to suggest primary subsistence strategies. Table 7.21 shows the frequency of lithic material (i.e.: core flakes, bifacially and unifacially modified tools, projectile points, and cores), and the presence or absence of groundstone and ceramics.

The artifact distribution suggests that the primary difference between assemblages is the presence or absence of ceramics and ground stone. Only one Archaic component contained ceramics, and this was identified as a single sherd of Alma Plain, an early Mogollon ware dated to AD 300 (Breternitz 1966). There is a slightly higher incidence of ground stone tools in Archaic assemblages, suggesting at least some xploitation of plant resources, and perhaps incipient horticulturalism. Comparison of the assemblages supports the general interpretation of Archaic lifeways as pre-ceramic and pre-horticultural.

Table 7.21. Archaic and Post Archaic Assemblage Raw Frequency Data. RAW GROUND MATERIALS* #FLAKES TOOLS POINTS CORES STONE CERAMICS BA/CA/CH/SL 10s 7 0 0 NO YES BA/CA/CH/QU 100s 10 0 4 YES NO BA/CH/QU 10s 1 0 1 YES NO CA/CH 10s 3 2 0 NO NO CA/CH/QU 100s 5 3 0 NO NO BA/CA/CH/OB/QU 10s 1 2 2 YES NO BA/CA/CH/OB/QU 10s 0 2 0 NO NO BA/CA/CH 10s 0 1 0 NO NO BA/CA/CH 10s 3 0 3 YES NO POST-ARCHAIC SITES BA/CH 10s 0 0 0 NO YES BA/CA/CH/OB/QU 10s 0 2 0 YES YES BA/CA/CH 100s 19 3 4 YES YES BA/CA/QU 10s 1 0 0 YES YES BA/CH 100s 2 0 0 YES YES BA/CH 10s 1 0 1 YES YES BA/CA/CH 10s 1 0 0 YES YES BA/CH 100s 12 5 0 YES YES BA/CA/CH 100s 4 8 0 YES YES BA/CA/CH/OB/QU 100s 3 4 0 YES YES BA/CA/CH/QU/SL 100s 15 7 0 YES YES *Raw Material Codes BA Basalt CA Chalcedony CH Chert OB Obsidian QU Quartzite SL Silicified Limestone Qualitative means of artifact distribution were computed from the assemblage frequency data, and are presented in Table 7.22. To test the statistical significance of the observed variation between Archaic and Post-Archaic assemblages, hypothesis testing was applied to quantitative data. Tests include t-Test, F-Test, and One-Way Analysis of Variance (ANOVA). Tables 7.23 and 7.24 show the results of the t-Test and F-Test, and Table 7.25 shows the results of the ANOVA. The results of all three tests suggests no statistically significant difference between the lithic tool assemblage distribution of Archaic and Post-Archaic components, suggesting that the importance of hunting did not diminish for area populations with the advent of horticulture during the Post-Archaic. Differences in the distribution of lithic raw material through time were determined using a presence/absence approach. Analysis of six raw material categories (basalt, chalcedony, chert, obsidian, quartzite, and silicified limestone) was carried out through testing the distributions using t-Test and ANOVA. Table 7.26 shows the results of analysis, which suggest no statistically significant difference between the frequency of raw materials found in Archaic and Post-Archaic period components. Table 7.22. Qualitative Assemblage Means of Archaic and Post-Archaic Sites. Archaic Sites ARTIFACT CATEGORY Flakes Tools Projectile Points Cores QUALITATIVE MEANS 10s 4 2 3 Post-Archaic Sites ARTIFACT CATEGORY Flakes Tools Projectile Points Cores QUALITATIVE MEANS 100s 7 3 1 Table 7.23. Archaic and Post-Archaic Assemblage Analysis: t-Test. t-Test Archaic Post Archaic Mean 6.4815 11.6212 Variance 353.4242 808.6696 Number 54 66 Pooled Variance 1927.2727 df 18 t -1.5006 P(T<=t) one-tail 0.0754 t Critical one-tail 1.7341 P(T<=t) two-tail 0.1508 t Critical two-tail 2.1009 Table 7.24. Archaic and Post-Archaic Assemblage Analysis: F-Test. F-Test Archaic Post Archaic Mean 6.4815 11.6212 Variance 353.4242 808.6696 Observations 54 66 df 53 65 F 2.2881 P(F<=f) one-tail 0.1099 F Critical one-tail 1.3957 Table 7.25. Archaic and Post-Archaic Assemblage Data: ANOVA Summary Statistics Groups Count Sum Average Variance Archaic 9 347 38.5556 1890.7778 Post-Archaic 11 767 69.7273 2950.8182 Source of Variation SS df MS F P-value F-crit Between Groups 8018.18 1 8018.18 3.4071 0.0798 4.3512 Within Groups 47066.90 20 2353.34 Total 55085.09 21 The Eco-Cultural Zonation Model The Model. A relationship between site location, chronology, slope direction, and number of features has been proposed for the prehistoric agricultural populations of the Mimbres area (Rice 1979) and the Tularosa Basin (Wimberly and Rogers 1977). Based on observations derived from field research and documented rainfall data (Martin and Plog 1973; Cordell 1979), Stuart and Gauthier have proposed a model for the Jornada Mogollon area which we have termed the Eco-Cultural Zonation Model (1988:185). The model suggests that populations affected by a winter-dominant rainfall pattern will select for site locations characterized by a south or southwest facing slope and a higher mean elevation, and will have a higher incidence of hearths. Summer rainfall pattern occupations will be characterized by an east or north facing slope, a lower mean elevation, and a fewer number of hearths (Stuart and Gauthier (1988:185; 216). Table 7.26. Archaic and Post-Archaic Assemblage Data: Lithic Raw Material. T-Test ARCHAIC POST-ARCHAIC Mean 1.64815 1.3484 Variance 3.1380 2.7228 Number 54 66 Pooled Variance 0.0864 df 8 t -1.5118 P(T<=t) one-tail 0.0845 t Critical one-tail 1.8595 P(T<=t) two-tail 0.1690 t Critical two-tail 2.3060 ANOVA Summary Statistics Groups Count Sum Average Variance Archaic 9 88 9.7777 13.6944 Post-Archaic 11 89 8.0909 25.4909 Source of Variation SS df MS F P-value F-crit Between Groups 0.0454 1 0.0454 0.0017 0.9670 4.3512 Within Groups 520.9090 20 26.0454 Total 520.9545 21 Chronology. Martin and Plog have documented a decline in effective rainfall for the central southwest region beginning around AD 870 (1973:51). This trend corresponds with a slight downhill shift in site location starting at about AD 900 in the Tularosa Basin (Wimberly and Rogers 1977). Stuart and Gauthier suggest that by AD 1000, as a response to the continued trend in summer-dominant precipitation, sedentary populations practicing horticulture would have moved site location significantly downhill in order to maximize their growing season (1988:217). Likewise, spring soil moisture would have been maximized by selecting for site locations near streams or alluvial fans, perhaps below east slopes, where the snowmelt above is likely to have occurred relatively late (1988:217). By AD 1100, with the summer-dominant pattern established, populations would have moved to basin floors in areas of major drainages. Around AD 1125, the amount of rainfall appears to have increased slightly, and the pattern of precipitation become temporarily more bimodal in distribution. At this time, the trend would have been to shift site locale upwards again. The model suggests that sites occupying the highest elevations would have bimodal dates of occupation, one between AD 850-900 and the other between AD 1200-1250 (1988:218). Maize The model also explores the relationship between the cultigens associated with sites at different elevational zonations and the intensity of cultivation. As microbotanical data indicative of the selected crops and the degree of agricultural intensification were not collected by survey, the analysis utilizes such relative horticultural correlates as the presence or absence of ground stone tools, ceramics, and thermally altered rock. Test of the Eco-Cultural Zonation Model Eco-cultural zonation is defined as the pattern resulting from the interrelationships of the environment (including temperature and precipitation regimes), site location, and site function which are predicted by the model. Test of the model will attempt to correlate environmental and cultural variables from data collected by survey. Elevation Elevation is identified as the primary ecological/environmental independent variable suggested by the model. The hypothesis predicts that differences of elevation will be significantly correlated with patterns of temporal affiliation and site orientation. Table 7.27 shows that the mean elevation for prehistoric sites is approximately 6175 feet, ranging from 5900 to 6900 feet, with a mode of 6400 and a standard deviation of 244.98 feet. Figure 7.33 shows that the frequency distribution of site elevations is characterized by two modal peaks, centered around 6000 feet and 6400 feet, respectively. The distribution suggests that, in the project area, Archaic sites tend to be located at higher elevations than sites of Mogollon and Mogollon/Anasazi affiliation. This pattern supports the expectations of the model, but correlation with local patterns of seasonal rainfall cannot be made as no rainfall data exist for the project area. The difference in elevational zonation between components of Archaic and Mogollon affiliation might also be interpreted as evidence of functional variation. Nomadic or semi-sedentary Archaic populations might be expected to select for site locational attributes which would increase the fitness of seasonal subsistence strategies primarily associated with hunting. These activities suggest selection for higher elevations which would provide good vantage points from which to watch for game, and valley views overlooking the drainage where game might be expected to gather. Sedentary, horticultural Mogollon populations might be expected to select site locations which would maximize the length of the growing season and which are in proximity to water. Table 7.27. Eco-Cultural Zonation Variables: Elevation. Site # Elevation Cultural Affiliation Mean St.Dev. N 107182 5960 Archaic 6154.44 219.04 9 107183 6020 Archaic 107194 5950 Archaic 107197 5900 Archaic 107201 6040 Archaic 107207 6400 Archaic 107208 6400 Archaic 107215 6400 Archaic 107245 6320 Archaic 107184 5900 Mogollon 5943.75 107.16 8 107185 5960 Mogollon 107186 6000 Mogollon 107189 6040 Mogollon 107190 5950 Mogollon 107191 6000 Mogollon 107192 5700 Mogollon 107204 6000 Mogollon 107193 6000 Mogollon/Anasazi 6053.33 110.33 3 107205 6180 Mogollon/Anasazi 107181 5980 Mogollon/Anasazi Slope The correlation between site slope direction (east/north and south/southwest) and elevation was tested using Pearson's and Spearman's correlation coefficients. Both indices produced a low correlation value. Pearson's Correlation Matrix yielded a value of 0.201. Spearman's Correlation Coefficient yielded a value of 0.193. These statistics suggest that prehistoric components in the project area are characterized by a low degree of correlation between site elevation and slope direction. Table 7.28 shows the summary statistics for these variables. Table 7.28. Slope and Elevation: Summary Data. E/N Slope S/SW Slope Measure Elevation Elevation Mean 6146.61 6226.15 Median 6020 6040 Standard Deviation 200.61 311.27 Range 500 1000 Minimum 5900 5900 Maximum 6400 6900 Count 23 14 Table 7.29. Correlation of Elevation and Slope. Regression Analysis Dependent Variable: Elevation of Prehistoric Sites N = 36 Variable Coefficient Std. Error Std. Coeff. T P(2-Tail) Constant 6000.778 120.394 0.000 49.843 0.000 Slope 133.702 86.962 0.255 1.537 0.133 Analysis of Variance Source Sum-of-Squares DF Mean Square F-Ratio P Regression 136554.124 1 136554.124 2.364 0.133 Residual 1964125.876 34 57768.408 To further test the expected relationship between elevation and slope, site locational data was subjected to regression analysis and analysis of variance. Table 7.29 shows the results of these analyses. The regression analysis suggests a correlation between the variables, and this implication is supported by the results of the analysis of variance. The results of these analyses suggest that a weak association between slope direction and elevation for prehistoric components in the project area. These observations tend to support the expectations of the model, and suggests that, in the project area, behavioral correlates exist between environmental and cultural variables that are reflected in the spatial patterning of prehistoric sites. Artifact Distribution and Elevation For this analysis, prehistoric components were grouped according to assemblage composition and elevational location. Artifacts selected for analysis are intended to suggest sedentism and the processing of plant resources, and include ceramics, ground stone, and fire-cracked rock/hearths. Table 7.30 shows the distribution of grouped components, their corresponding elevations, and the presence or absence of the selected artifacts. It should be noted that the size of the sample is very small, thus weakening the degree of significance of the statistical tests. Five Archaic components are characterized by the inclusion of one or more of these artifact classes in their assemblage. Correlation of these components with their elevational location suggests that a higher frequency of the selected artifacts exist at lower elevations than at higher elevations. Ground stone, thermally altered rock/hearths and the single ceramic sherd, are present in all the components located at the lower elevations. Components located at higher elevations appear to lack one or more of the selected artifacts. Although this was the largest sample (n=9), the variation in frequency distribution is slight, and the statistical significance is low. Four groups of Early Pithouse to Late Pueblo Jornada Mogollon components included the selected artifact classes. Three areas of elevational zonation were defined for these groups. The "middle" elevational zone is characterized by mean elevations only 50m apart; in the "low" elevational zone, the mean is 100m apart from the middle zone; and in the "high" elevational zone the mean is 900m apart from the middle zone. The grouped components are characterized by very low frequencies of the selected artifacts at high (11%) and low (16%) elevations, and a very high frequency of the selected artifacts at middle elevations (72%). The small sample (n=8) suggests the observed variation is low in statistical significance. Three mixed Jornada Mogollon and Anasazi components contained the selected artifact classes. These components show a higher frequency of the selected artifacts in the low elevational zones, and a lower frequency of selected artifact categories in the high elevational zones. It should be noted that this constituted the smallest sample (n=3). The observed variation is slight, and the statistical significance of the patterns observed is very weak. Evaluation of the Model The analyses suggest that, in the absence of microbotanical and other data which are generally not collected in pedestrian surveys, elevational data appears to be a useful independent variable in determining temporal patterns of eco-cultural zonation. Slope direction, although a weaker correlation, was also shown to be useful in predicting temporal variation of site location. These variables suggest potentially useful correlations with site functional variation and the presence or absence of certain artifact classes from the assemblage composition. Although direct correlations between site function, artifact categories, and site location has been suggested by the data, it must be noted that the samples used in this analysis are too small to indicate robust statistical significance. Future research in the area of the present project is expected to augment the data set and provide further corroboration of the utility of this model. Table 7.30. Eco-Cultural Zonation: Grouped Prehistoric Components. Grouped Groundstone Ceramics FCR Combined Grouped Cultural Affiliation #Sites Yes/No #Sites Yes/No #Sites Yes/No #Sites Yes/No Elevation Archaic 1 0 1 2 5900 Archaic 2 0 0 2 5950 Archaic 1 1 0 2 6000 Archaic 1 0 0 1 6300 Archaic 0 0 0 0 6400 Mogollon 1 1 1 3 5900 Mogollon 2 2 0 4 5950 Mogollon 2 4 3 9 6000 Mogollon 1 1 0 2 6900 Mogo/Ana 1 1 0 2 5950 Mogo/Ana 1 1 0 2 6000 Mogo/Ana 0 1 0 1 6150 Totals 13 12 5 30