Frontiers in Colorado Paleoindian Archaeology: From the Dent Site to the Rocky Mountains

7. Building a Picture of the Landscape Using Close-Interval Pollen Sampling and Archaeoclimatic Modeling: An Example from the KibRidge-Yampa Paleoindian Site, Northwestern Colorado

195 Linda Scott Cummings, R. A. Varney, and Reid A. Bryson C h a p t e r s e v e n Building a Picture of the Landscape Using Close-Interval Pollen Sampling and Archaeoclimatic Modeling An Example from the KibRidge-Yampa Paleoindian Site, Northwestern Colorado Understanding the past environment is made more difficult because no modern analogs exist for many of the previous vegetation communities or environmental systems. Vegetation communities when Paleoindians lived on the North American continent were governed by climatic conditions and an earth-sun relationship that do not exist on earth today. Therefore, rather than simply examining pollen records to provide information concerning the plants that were present, we have found it more insightful to incorporate archaeoclimatic modeling to understand the variations in seasonal temperature and precipitation that affected local plants and animals. Once the model is created, it is essential that pollen or other proxy records be consulted to “ground truth” the model. That is to say, models are excellent tools, but they should not be considered a substitute for data. Examining data at the closest interval possible provides considerably more Linda Scott Cummings, R. A. Varney, and Reid A. Bryson 196 information concerning local vegetation than does examining data averaged over long periods of time. RELEVANCE OF MODELS AND MICRO-BOTANIC RECORDS TO PEOPLE The “average” life span of a human living during prehistory has often been calculated at 35 years. This does not suggest that the average person lived to age 35 but rather that when infant mortality and other factors often leading to an early death were figured into the equation, the calculated expected life span was approximately 35 years. People who survived early childhood had a good chance of living into their 50s or perhaps longer, establishing what we will use as a life span for purposes of this discussion. Decisions on where to live were typically made on seasonal, annual, or decadal scales. Animal migrations are often seasonal. Plants are annuals, biennials, or perennials, with shrubs and some trees living 50 to several hundred years. These comparisons are necessary to establish a good sampling interval to understand the paleoenvironment. If decisions were made over relatively short periods of time (seasonally, annually, or perhaps decadally), then examination of the paleoenvironmental record every 200–500 years seems very inappropriate. Previous pollen studies were considered sufficient if an individual sample represented a few hundred years. In fact, the vast majority of stratigraphic pollen and phytolith analyses for North America are based on a sampling interval that yields data every few hundred years, although bogs and lakes yield samples at 100year intervals more often than sediment columns do. A survey of pollen records depicted in Pollen Records of Late-Quaternary North American Sediments (Bryant and Holloway 1985) displays sampling intervals ranging from a hundred to several hundred or even a thousand years for most of the records. The purpose in examining these records was to establish trends in vegetation representing paleoenvironmental conditions, usually documenting retreat of the glaciers or perhaps onset or conclusion of the middle Holocene period. Other studies referenced in that survey were conducted to provide general information concerning local and regional vegetation as part of the paleoenvironmental interpretation of the landscape used as the setting for human populations. Much of this work has assumed either relatively stable vegetation communities through time or gradual changes in vegetation communities, so sampling interval was assigned little importance. Rather, sampling intervals of 10 cm were adopted as convenient. Although pollen sampling of very short vertical intervals (1 to 2 cm), separated by sediments not sampled, has been recommended for many years (Scott 1980), the recommendation has rarely been followed. Spacing samples at 5 to 10 cm intervals in a stratigraphic column has been the norm in many studies to accommodate small budgets for paleoenvironmental work. Unfortunately, this has often resulted in archaeologists collecting an entire block of 5 cm or 10 cm cubes, averaging large quantities of sediments, rather than collecting discrete 1 or 2 cm layers and then selecting samples 5 to 10 cm apart for analysis. One of Building a Picture of the Landscape Using Close-Interval Pollen Sampling 197 the shortcomings of small budgets for laying a foundation for the archaeological record is that they do not recognize or give importance to understanding variations in the past environment, which provides a baseline for all life in the area. Plant...