Early Commuting
Exploring the mobility of prehistoric people
By Fumiyasu Arakawa and Christopher Nicholson

Although conventional wisdom holds that the shortest distance between two points is a straight line, it is rarely the easiest way to get there. Whether traveling to the grocery store, to work, or even to the mailbox, people inevitably face physical obstacles. This is as true in the past as it is today.

A cost-weight map for a habitation site to the nearest toolstone quarry

To better understand territoriality and social interaction in an ancient society, students and faculty in the Anthropology Department at Washington State University and researchers at Crow Canyon Archaeological Center, a not-for-profit organization, used GIS to examine how prehistoric people in the Mesa Verde region of southwestern Colorado traveled across the landscape 1,000 years ago as they collected the stones they used to make tools.

This GIS study is only one component of the Village Ecodynamics Project undertaken by Washington State University and Crow Canyon Archaeological Center. The project was designed to help archaeologists and anthropologists better understand the factors that influence settlement patterns of small-scale agrarian peoples. Participants in this study of sociopolitical organization and prehistoric technology chose a modern technology—GIS—that allows them to examine spatial relationships in addition to environmental variables.

Because people utilize a given landscape in a unique manner based on the area's environmental constraints and obstacles, researchers decided to model how these prehistoric people moved across a landscape based on a very simple set of environmental parameters. Using ArcGIS 9.1 with the ArcGIS Spatial Analyst extension, they examined social interactions by investigating how people did something as simple as walk over the hill to collect rocks.

Mesa Verde and Its People

Habitation site markers scaled by energy expenditure. The larger circles indicate higher energy expenditures for gathering toolstone material.

During the long prehistory of Mesa Verde, its people transitioned from hunters and gatherers to small-scale horticulturists to full-time sedentary farmers. Some lived in cliff dwellings, while others lived in villages on the top of mesas. This all changed suddenly. Most of the area's population disappeared in the late 1200s. Until this time, people relied heavily on stone tools for hunting, farming, and butchering animals. However, getting materials to make good stone tools was not easy. Since getting materials to make toolstones required effort, researchers asked if toolstone procurement (i.e., collecting stones to make tools) could also be an indicator of territoriality and/or other social interactions.

Analyzing Mobility

ArcGIS 9.1 was used for four activities:

• Building a reclassified cost raster for the study area that simulates the terrain traveled
• Implementing a cost-weight analysis from each habitation site to the nearest quarry
• Using the kriging program to generate the energy-expenditure values across the study area for each time period
• Creating a difference map based on the results of kriging maps by subtracting the values in the later periods from the next earlier period using map algebra

First, a high-resolution digital elevation model (DEM) raster map of the region was reclassified by slope. This analysis of the terrain traveled by humans in the canyon environment around Mesa Verde was based on the assumption that steeper slopes made travel more difficult.

Interpolated maps of energy expenditure using the kriging function for four time periods

Next, a cost-weight analysis was done to calculate the approximate energy expenditure of people traveling from their homes to a toolstone quarry. Cost-distance values reflect the walking costs in a canyon landscape. In the model, habitations were added as beginning points and quarries were added as destination points. The Extract Values to Point tool, available with the ArcGIS Spatial Analyst extension, was used to approximate the cost of walking to get toolstones for each habitation.

Looking at Energy Expenditure

To visualize how energy-expenditure values are spread across this landscape, the kriging function in Spatial Analyst was used to create maps based on the energy-expenditure values when procuring toolstone materials for several different time periods. This information was thematically mapped. Dark colors indicated high-energy expenditures. White indicates low-energy expenditure. For example, during the early Pueblo II period (A.D. 920�1060), no dark areas are shown on the map. This suggests that early Pueblo II residents expended less energy when procuring toolstones. In contrast, energy expenditure values for the late Pueblo III period (A.D. 1225�1280) indicate that residents in this era appear to have expended far more energy performing a similar task.

A difference map showing areas where the energy expenditure values for acquiring toolstones increased or decreased in late Pueblo III sites relative to early Pueblo III sites

A series of ArcGIS maps was created to identify areas where energy expenditures by central Mesa Verde inhabitants changed from one time period to another. In ArcGIS Spatial Analyst, the Raster Calculation tool was used to subtract energy expenditure values of a later period from an earlier period. Mapping the difference in values between these datasets visualized the changes in expended energy from one period to the next. Yellow-shaded areas indicate where the early Pueblo III inhabitants expended more energy. Red shading identifies similar areas for late Pueblo III residents. The difference map illustrated where people expended either more or less energy for acquiring raw materials.

Cases in which more energy was expended during a given time period at a given location may indicate the development of territoriality in Puebloan communities. This analysis also suggests that, because of an increased population during the early Pueblo III period, people were forced to expend more energy when seeking toolstone materials. This may have meant that people had to travel farther from home as part of their work making them early long-distance commuters.

Conclusion

Using ArcGIS allowed researchers to see how physical constraints (e.g., in this case, topographic challenges) affected toolstone acquisition activities of the residents of Puebloan communities. Cost-weight and kriging maps displayed how energy was expended when getting materials for making tools. The authors strongly suggest that archaeologists can implement similar GIS procedures to understand and reconstruct human mobility (using energy expenditures) and social interactions using the simplest of environmental parameters along with some type of currency (toolstones, in this example) that humans need and actively seek.

About the Authors

Dr. Fumiyasu Arakawa (FArakawa@crowcanyon.org) is a laboratory analysis specialist at the Crow Canyon Archaeological Center in Cortez, Colorado, and received a doctorate in anthropology from Washington State University. Christopher Nicholson (cnichol5@uwyo.edu) is the outreach and technology coordinator for the Water Resources Data System at the University of Wyoming in Laramie, Wyoming. He holds a master's degree in anthropology from Washington State University.

For More Information

The instructor-led course, Working with ArcGIS Spatial Analyst, is available. Visit www.esri.com/training for more information about this course.