Search

Between 2100 and 500 BP, maize farming spread throughout the American Southwest, driving changes in land use patterns and relative increases in the number of archaeological sites, which is commonly interpreted as increases in population. In the context of the Southwest, with its variable topography and limited water sources, the locations suitable for productive maize agriculture are limited. Higher elevation locations often have a greater abundance of water, both in the form of precipitation and perennial water sources but lack the growing season lengths required for maize farming. Lower elevation locations have the long growing seasons required for maize, but often have higher evaporation rates corresponding with higher temperatures, and thus decreased soil moisture retention. The relationship between elevation and maize agriculture is well recognized, and locations where both growing season length and moisture requirements can be met serve as a Goldilocks Zone in which maize farming is possible. Here, we use a regional approach to model the Goldilocks Zone for maize farming in Utah, and then contextualize patterns of settlement and abandonment on the West Tavaputs Plateau (WTP) in east-central Utah. We use archaeological sites with direct evidence of maize cultivation, modern growing degree days estimates, and modern estimates of moisture abundance to bracket the prehistoric maize farming niche in Utah. Then, we apply the model to archaeological data from the WTP to test hypotheses derived from the Ideal Free Distribution model about the effects population and precipitation changes had on prehistoric land use patterns of settlement and abandonment, and whether the patterns of settlement and abandonment differ from one another. We find that a) population drives land use patterns in the WTP, but not in the way we originally hypothesized, suggesting there is an important decision variable missing from our model, which may pertain to the specific topographic setting of the WTP and the distribution of its water sources, b) precipitation has no effect on land use patterns suggesting that maize agriculture on the WTP was dependent on irrigation, and c) that patterns of settlement and abandonment seem to mirror each other. [ABSTRACT FROM AUTHOR]

Climatic change appears to influence major patterns in human history. However, confirming the association between climatic events and adaptive human responses is not straightforward given the potential for spurious correlations and uncertainty in the timing of events. Here we leverage theory from behavioral and population ecology to generate predictions about causal relationships between climate and settlement decisions. We then test those predictions using statistical methods designed to evaluate associations under uncertainty. This allows us to evaluate a comprehensive record of radiocarbon dated archaeological sites in the Basin-Plateau region of North America relative to a record of effective precipitation reconstructed using Macrophysical Climate Models. Consistent with theory, the results show that regional populations declined in the mid-Holocene during a prolonged period of reduced precipitation, leading to the abandonment of many major rockshelter sites. Remaining populations concentrated at open-air sites around well-watered sand dune localities where they constructed pithouses and adopted more intensive subsistence practices. These findings illustrate an adaptive response of hunter-gatherer populations to past climate change that supports theoretical predictions and is unlikely to result from spurious correlations or uncertainties in the timing of events. [Display omitted] [ABSTRACT FROM AUTHOR]

Predictive models are central to both archaeological research and cultural resource management. Yet, archaeological applications of predictive models are often insufficient due to small training data sets, inadequate statistical techniques, and a lack of theoretical insight to explain the responses of past land use to predictor variables. Here we address these critiques and evaluate the predictive power of four statistical approaches widely used in ecological modeling—generalized linear models, generalized additive models, maximum entropy, and random forests—to predict the locations of Formative Period (2100–650 BP) archaeological sites in the Grand Staircase-Escalante National Monument. We assess each modeling approach using a threshold-independent measure, the area under the curve (AUC), and threshold-dependent measures, like the true skill statistic. We find that the majority of the modeling approaches struggle with archaeological datasets due to the frequent lack of true-absence locations, which violates model assumptions of generalized linear models, generalized additive models, and random forests, as well as measures of their predictive power (AUC). Maximum entropy is the only method tested here which is capable of utilizing pseudo-absence points (inferred absence data based on known presence data) and controlling for a non-representative sampling of the landscape, thus making maximum entropy the best modeling approach for common archaeological data when the goal is prediction. Regression-based approaches may be more applicable when prediction is not the goal, given their grounding in well-established statistical theory. Random forests, while the most powerful, is not applicable to archaeological data except in the rare case where true-absence data exist. Our results have significant implications for the application of predictive models by archaeologists for research and conservation purposes and highlight the importance of understanding model assumptions. [ABSTRACT FROM AUTHOR]

Abstract: One or more of three intense and persistent droughts impacted some Native American cultures in the early-11th, middle-12th and late-13th centuries, including the Anasazi, Fremont, Lovelock, and Mississippian (Cahokian) prehistorical cultures. Tree-ring-based reconstructions of precipitation and temperature indicate that warm drought periods occurred between AD 990 and 1060, AD 1135 and 1170, and AD 1276 and 1297. These droughts occurred during minima in the Pacific Decadal Oscillation and may have been associated with positive values of the Atlantic Multidecadal Oscillation. Each of the Native American cultures was supported, to a greater or lesser degree, by precipitation-dependent resources. Both the Four Corners region and Cahokia were sites of intense growth between about AD 1050 and 1130, and by AD 1150, cultures in both regions were undergoing stress. By AD 1300 the Anasazi and Fremont cultures had collapsed and their residual populations had either left their homelands or withered. In the case of Fremont populations, the AD 990–1060 drought may have had the greatest impact. This drought also may have affected the Anasazi, for it was at the end of this drought that some people from Chaco migrated to the San Juan River valley and founded the Salmon Ruin great house. Detailed data do not exist on the number of Lovelock habitation sites or populations over time; however, Lovelock populations appear to have retreated from the western Great Basin to California by AD 1300 or shortly thereafter. [Copyright &y& Elsevier]