By Brian Tomaszewski, Ph.D. Candidate, Department of Geography, the Pennsylvania State University

Surface artifact densities at the Calixtlahuaca site collected with GPS (map by Juliana Novic, Arizona State University).

Archaeological investigation often begins with systematic field surveys to inventory cultural features and artifacts visible on the ground surface of a particular site or region. These inventories are then used, in part, to gain insight into the overall spatial and temporal dimensions of a site based on artifact types found and to make decisions about which areas within a site will receive priority when conducting traditional ground excavations. Determining which areas to excavate is important not only for the underlying research questions of a given project but also for more pragmatic reasons, as most archaeological investigations are limited by financial and temporal constraints. It is with these constraints in mind that a cost-effective approach to integrating GPS technology with GIS was developed for an archaeological field survey in central Mexico. This approach allowed for the collection of accurate, reliable information about archaeological ground conditions, such as surface artifact densities, subsurface prehistoric structures, and natural features that could be integrated into and visualized with GIS for excavation prioritization and planning.

The field survey using this approach was conducted at the Aztec-period Calixtlahuaca site, located in the Toluca Valley approximately 40 miles (60 kilometers) due west of Mexico City. The earliest known occupancy of the site dates to approximately 1100 A.D. The site was occupied by indigenous peoples until its conquest by the Spanish in the late 15th century. The site has several pyramid or temple structures that were excavated in the 1930s by Mexican authorities.

Earlier archaeological investigations at Calixtlahuaca focused on temples and tombs. Today, archaeologists are looking for evidence of where nonelite residents lived on the site.

The team of investigators that conducted the survey are part of a National Science Foundation (NSF) grant to examine the nonelite residential occupancy of the site, a research approach that purposely de-emphasizes the investigation of so-called "monumental" structures, such as pyramids, temples, and tombs. A 4.2-square-kilometer systematic survey was conducted to find clues as to where houses and other nonelite structures were located, and the results of the survey will be used to prioritize which areas will be excavated in the following field season.

ArcGIS Desktop (ArcView) was at the core of the geospatial solution developed for the field survey. In addition, Garmin eTrex Legend GPS devices were used for field data collection. These devices are easy to use and rugged; have large storage capacity; were accurate for the spatial needs of the field survey (within 1 to 5 meters at most control points when tested); and, at a cost of less than $200 per unit, were within the project budget. In addition, the free Minnesota Department of Natural Resources (DNR) Garmin Extension for ArcView facilitated eTrex Legend's GIS functionality.

This DNR program provided the key link to integrating GPS-collected data with ArcView. The program served as a bridge between the Garmin device and the GIS and allowed waypoints to be uploaded and downloaded from the GPS device, exported raw GPS data to shapefile and geodatabase feature classes, and provided functionality for synchronizing digital photographs with GPS waypoints to create hot link features.

All data collected was stored in a personal geodatabase, which was selected because of its native Microsoft Access file format (.mdb). This allowed spatial data to be integrated with nonspatial datasets that were developed in a standard Access database. Access data entry forms were quickly developed on personal geodatabase tables without any concern of damaging the spatial component of the data as spatial fields were simply left out of data entry interfaces. In addition, nonspatial data could be quickly and easily visualized in the GIS using join operations onto spatial data without any interoperability issues arising from trying to join data in different file formats.

Surface artifacts collected and mapped in the field provide insight into where subsurface structures and other cultural features may lie.

The ArcView functionality proved essential to the project. An important data source used in the field surveys was digital orthophotos purchased from the state of Mexico. When the projection of these photos was found to not correspond with ground-truth data collected by the GPS devices, georeferencing functions of the ArcGIS Spatial Analyst extension were used to adjust the imagery, which after rectification and warping then matched correctly with the GPS data. Because of budgetary constraints, the crews used hard-copy maps in the field to mark relevant information. A critical component of these maps was Universal Transverse Mercator (UTM) coordinate grids displayed at multiple intervals. Field crews matched the coordinates from these grids shown on the paper maps with their real-time position coordinates shown in the GPS. This allowed the crews to stay oriented in the field and not waste time continually trying to figure out their position from the paper maps alone. A custom ArcObjects script was created to generate grid lines at user-defined intervals in a grid feature class layer that was used to generate these maps.

In conclusion, the use of ArcGIS Desktop coupled with a cost-effective GPS/GIS solution, ease of use, and integration capabilities of geodatabase technology allowed for the rapid development of a comprehensive geospatial solution to meet the needs of an archaeological field survey. By using this approach, financial and temporal constraints of the project were overcome without compromising the scientific integrity of the overall research goals. Although applied to archaeology, this type of approach could be used by any researcher who needs an integrated GIS/GPS/data solution but lacks the financial resources for an optimal solution.

About the Author

Brian Tomaszewski, a Ph.D. candidate and graduate research associate in the Department of Geography and GeoVISTA center at the Pennsylvania State University, has an M.A. degree in geography from the State University of New York, Buffalo. His research interests include GIScience, geocollaboration, historical GIS, and crisis management.

More Information

For more information on this archaeological project, visit www.public.asu.edu/~mesmith9/Calix. For more information on the geospatial methods used on the project, contact Brian Tomaszewski, Ph.D. candidate, Department of Geography, the Pennsylvania State University (e-mail: bmt139@psu.edu). For more information on the DNR Garmin Extension for ArcView program, visit www.dnr.state.mn.us/mis/gis/tools/arcview/extensions/DNRGarmin/DNRGarmin.html.

Acknowledgments

Funding for this project comes from NSF grant #0618462, Urbanization and Empire at the Aztec-Period Site of Calixtlahuaca, Dr. Michael E. Smith, principal investigator. Additional funding was provided by Arizona State University. Esri software was provided through the Esri campuswide site license program (www.esri.com/campuswidelicense) at Arizona State University. The project also thanks the Instituto Nacional de Antropolog�a e Historia (INAH) and the state of Mexico for their cooperation.