Archaeology and Geoinformatics: Case Studies from the Caribbean

Archaeology and Geoinformatics

CARIBBEAN ARCHAEOLOGY AND ETHNOHISTORY

Series Editor Antonio L. Curet

Archaeology and Geoinformatics

Case Studies from the Caribbean

EDITED BY BASIL A. REID

THE UNIVERSITY OF ALABAMA PRESS

Tuscaloosa

Copyright © 2008

The University of Alabama Press

Tuscaloosa, Alabama 35487-0380

All rights reserved

Manufactured in the United States of America

Typeface: ACaslon

∞

The paper on which this book is printed meets the minimum requirements of American National Standard for Information Sciences-Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984.

Library of Congress Cataloging-in-Publication Data

Archaeology and geoinformatics : case studies from the Caribbean / [edited by] Basil A. Reid.

           p.   cm. — (Caribbean archaeology and ethnohistory)

    Includes bibliographical references and index.

    ISBN 978-0-8173-1601-3 (cloth : alk. paper) — ISBN 978-0-8173-5470-1 (pbk. : alk. paper) — ISBN 978-0-8173-8053-3 (electronic) 1. Indians of the West Indies—Antiquities. 2. Archaeological geology—Caribbean Area. 3. Excavations (Archaeology)—Geographic information systems—Caribbean Area. 4. Caribbean Area—Antiquities. I. Reid, Basil A., 1961–

    F1619.A73 2008

    972.900285—dc22

2007036716

Contents

List of Illustrations

Acknowledgments

Introduction

Archaeology and Geoinformatics: Case Studies from the Caribbean

Basil A. Reid

PART I: ARCHAEOLOGY, GIS, AND VISIBILITY MODELS

1. The Caribbean: A Continent Divided by Water

Joshua M. Torres and Reniel Rodríguez Ramos

PART II: ARCHAEOLOGY, GIS, AND CULTURAL RESOURCE MANAGEMENT

2. Developing Weights-of-Evidence Predictive Models for the Cultural Resource Management of Pre-Columbian Sites in Trinidad

Basil A. Reid

3. Forward Planning: The Utilization of GIS in the Management of Archaeological Resources in Barbados

Kevin Farmer

4. Developing an Archaeological Information System for Trinidad and Tobago

Bheshem Ramlal and Basil A. Reid

PART III: ARCHAEOLOGY, GIS, CARTOGRAPHY, GPS, SATELLITE IMAGERY, AERIAL PHOTOGRAPHY, AND PHOTOGRAMMETRY

5. Maps, Matricals, and Material Remains: An Archaeological GIS of Late-Eighteenth-Century Historic Sites on St. John, Danish West Indies

Douglas V. Armstrong, Mark W. Hauser, David W. Knight, and Stephan Lenik

6. Understanding Nevis: GPS and Archaeological Field Survey in a Postcolonial Landscape

Roger H. Leech

7. The Use of Imagery to Locate Taino Sites in Jamaica in a GIS Environment

Parris Lyew-Ayee and Ivor Conolley

PART IV: ARCHAEOLOGY AND GEOPHYSICS

8. Geophysics and the Search for Raleigh’s Outpost on Trinidad

Eric Klingelhofer

9. Geophysics and Volcanic Islands: Resistivity and Gradiometry on St. Eustatius

R. Grant Gilmore III

Conclusion

Postscript: Archaeology and Geoinformatics from a Caribbeanist Perspective

Basil A. Reid

Glossary of Terms

References Cited

Contributors

Index

Illustrations

FIGURES

1.1. Map of the Caribbean Basin with GTOPO30 digital elevation model (DEM) and political boundaries

1.2. Recalculated DEM using Bowditch’s standard distance to the visible horizon formula

1.3. Recalculated DEM using Bowditch’s standard distance to the visible horizon formula showing sample points from/to which visibility is possible

1.4. Visibility ranges based on point sample

1.5. Map showing variability in distances between coasts and landmass visibility ranges

1.6. Overlapping visibility areas in the Greater Antilles

1.7. Overlapping visibility areas in the Lesser Antilles

2.1. Map of Trinidad showing the location of the three watersheds selected for this study

2.2. Map of Trinidad showing the location of the watersheds in relation to the island’s physiography

2.3. Map of Trinidad and Tobago showing a sample of pre-Columbian sites on the island

2.4. Flow chart of weights-of-evidence procedure used in this study

2.5. Output weights of Landform evidential theme (Cipero watershed)

2.6. Output weights of Soil Texture evidential theme (Cipero watershed)

2.7. Output weights of Land Capability evidential theme (Cipero watershed)

2.8. Output weights of Relief evidential theme (Cipero watershed)

2.9. Output weights of Landform evidential theme (South Oropouche watershed)

2.10. Output weights of Land Capability evidential theme (South Oropouche watershed)

2.11. Output weights of Relief evidential theme (South Oropouche watershed)

2.12. Output weights of Landform evidential theme (Rest North watershed)

2.13. Output weights of Soil Texture evidential theme (Rest North watershed)

2.14. Output weights of Land Capability evidential theme (Rest North watershed)

2.15. Output weights of Relief evidential theme (Rest North watershed)

2.16. Posterior probability of Cipero watershed

2.17. Posterior probability of South Oropouche watershed

2.18. Posterior probability of Rest North watershed

2.19. Archaeological site favorability for Cipero watershed

2.20. Archaeological site favorability for South Oropouche watershed

2.21. Archaeological site favorability for Rest North watershed

3.1. Map of Barbados showing prehistoric site locations

4.1. Entity-relationship diagram of conceptual model

4.2. Attribute data entry form

4.3. Map of Trinidad created using the archaeological sites data set

4.4. Map of Trinidad created using the land-use data set

4.5. Selected sites that are within 500 m of a river and/or the coastline

4.6. Sites intersected with digital elevation model

5.1. Oxholm’s map of St. John, 1780

5.2. Oxholm’s map of St. John, 1800

5.3. Beck’s map of St. Croix, 1754

5.4. Map of Coral Harbor, 1720

5.5. Oxholm’s 1777 draft map of St. John

5.6. Fortification of Cruz Bay (Oxholm drawing)

5.7. Fort at Coral Bay (Oxholm drawing)

5.8. Color aerial survey/base map, 1994

5.9. Field survey: Mark Hauser using GPS to locate sites

5.10. Identification of historic sites (Cinnamon Bay Plantation)

5.11. GIS plot of sites dating to 1780 and 1800 (quarter boundaries also shown)

5.12. Plantation production by estate and quarter (sugar, cotton, provision), 1780 versus 1800

5.13. Change in distribution of unfree (enslaved) population from 1780 to 1800

6.1. Map of Nevis showing the divisions

6.2. Mountravers: the plantation landscape

6.3. Upper Woodland plantation, plans from archaeological survey

6.4. Parris’s Garden, an eighteenth-century villa

7.1. Map of the parish of Trelawny located in north-central Jamaica

7.2. An example of how the X-parallax is observed

7.3. An infrared image of the study area

7.4. Land-cover classifications of the study area

7.5. Map showing potential areas for future Taino site exploration

8.1. Locational map of Los Gallos Point, Trinidad

8.2. Site map of Los Gallos Point, also depicting the possibly contemporary aboriginal Quemada Point site

8.3. Mercer team employing magnetometer at Los Gallos Point in 1998

8.4. Plan of the 1998 magnetometer surveys

8.5. Mercer team using Mark 1 prototype resistivity meter in 2002

8.6. Magnetometer survey (1998) of the Main Survey Area, with the smaller 2002 resistivity survey area imposed

8.7. Detail of the 1998 magnetometer survey with the same area of the 2002 resistivity survey

8.8. Mercer team taking soil temperature readings in 2002

8.9. Test Unit 9, 2004

8.10. Test Unit 10, 2004; east–west soil stain of possible slot

9.1. Geographical location of St. Eustatius

9.2. Location of Pleasures Estate Plantation and English Quarter Plantation on St. Eustatius

9.3. Several distinct foundation areas evident in the resistivity meter results for English Quarter Plantation

9.4. English Quarter Plantation as depicted on the 1781 P. F. Martin map; the Pleasures Estate Plantation as depicted on the same map

9.5. Geophysics identified additional structural remains at the Pleasures Estate Plantation

TABLES

2.1. Sites in Cipero watershed

2.2. Sites in South Oropouche watershed

2.3. Sites in Rest North watershed

2.4. Details of the analysis parameters of selected watersheds

2.5. Attributes of output weights of Landform evidential theme (Cipero)

2.6. Attributes of output weights of Soil Texture evidential theme (Cipero)

2.7. Attributes of output weights of Land Capability evidential theme (Cipero)

2.8. Attributes of output weights of Relief evidential theme (Cipero)

2.9. Attributes of output weights of Landform evidential theme (South Oropouche)

2.10. Attributes of output weights of Soil Texture evidential theme (South Oropouche)

2.11. Attributes of output weights of Land Capability evidential theme (South Oropouche)

2.12. Attributes of output weights of Relief evidential theme (South Oropouche)

2.13. Attributes of output weights of Landform evidential theme (Rest North)

2.14. Attributes of output weights of Soil Texture evidential theme (Rest North)

2.15. Attributes of output weights of Land Capability evidential theme (Rest North)

2.16. Attributes of output weights of Relief evidential theme (Rest North)

2.17. Attributes of weights-of-evidence (Cipero)

2.18. Attributes of response theme (Cipero)

2.19. Attributes of posterior probability (Cipero)

2.20. Attributes of conditional independence (Cipero)

2.21. Attributes of weights-of-evidence (South Oropouche)

2.22. Attributes of response theme (South Oropouche)

2.23. Attributes of posterior probability (South Oropouche)

2.24. Attributes of conditional independence (South Oropouche)

2.25. Attributes of weights-of-evidence (Rest North)

2.26. Attributes of response theme (Rest North)

2.27. Attributes of posterior probability (Rest North)

2.28. Attributes of conditional independence (Rest North)

2.29. Archaeological site predictive model for Cipero watershed

2.30. Archaeological site predictive model for South Oropouche watershed

2.31. Archaeological site predictive model for Rest North watershed

4.1. Major applications of the AIS

4.2. Data required for the AIS

4.3. Major attributes stored for archaeological sites

7.1. Land-cover characteristics of the study region

9.1. Relative shipping activity in colonial ports

Acknowledgments

I wish to thank all the authors of the various chapters for their stellar contributions. It was indeed a pleasure working with them. Their essays, which reflect tremendous insight and an excellent grasp of the subject matter, are well appreciated and will, no doubt, serve to develop a better appreciation of geoinformatics within the context of Caribbean archaeology. I wish to express my profound gratitude to Narissa Seegulam, Daren Dhoray, Tariq Khan, John De Sormeaux, Kerry Bullock, Chris Riley, Naseema Hosein, Maria Peter-Joseph, Lystra Baksh-Moti, Marcela Boatswain, Serene Joseph, Kimika Lai Tan, Luke Mahase, and Dale Flemming for assisting with various aspects of the project such as typing, collating, and enhancing maps, figures, and tables coupled with compiling both the glossary and references in the volume. Heather Cateau provided useful information on references relating to West Indian plantation societies. Her contribution is gratefully acknowledged. I am particularly grateful to my wife, Joan, and our son, Gavin, for their moral support and incomparable patience as I sat up long hours in the nights vetting and editing the various chapters. Finally, thanks to the staff of the University of Alabama Press for their encouragement at every stage of this publication.

Basil A. Reid

Introduction

Archaeology and Geoinformatics: Case Studies from the Caribbean

Basil A. Reid

This volume, Archaeology and Geoinformatics: Case Studies from the Caribbean, presents a miscellany of both interesting and informative essays on the use of geoinformatics in Caribbean archaeology. The contributions are based on case studies drawn from specific island territories, namely, Barbados, St. John, Jamaica, Nevis, St. Eustatius, Puerto Rico and Trinidad and Tobago, with Chapter 1, which focuses on interisland interaction and landscape conceptualization in the Caribbean region, being the exception. Geoinformatics is one of the relatively new emphases in archaeology and can be defined as an interdisciplinary field that develops and uses information science and science infrastructure to address the problems of geosciences. In order to achieve its objectives, geoinformatics employs a battery of integrative and innovative approaches in analyzing, modeling, and developing extensive and diverse data sets. Several disciplines fall within the general purview of geoinformatics, namely, geographic information systems (GIS), global positioning systems (GPS), satellite imagery, aerial photography, photogrammetry, cartography, and geophysical surveys. However, while these techniques are increasingly being utilized in archaeology (e.g., Allen et al. 1990; Brophy and Cowley 2005; Carr 1982; Clarke 1990; Knowles 2002; Lock and Stancic 1995), most publications largely revolve around North American and European case studies, with scant attention being paid to the Caribbean. By demonstrating that this region—like anywhere else in the world—is fertile ground for the application of geoinformatics in archaeology, this volume places a well-needed scholarly spotlight on the Caribbean.

THE USEFULNESS OF GEOINFORMATICS IN CARIBBEAN ARCHAEOLOGY

The Caribbean is fertile ground for the following reasons. To begin with, the hive of archaeological activity in the region since the 1980s has led to an increasing demand for state-of-the-art technologies. Fourteen years ago, William Keegan (1994:255) exclaimed that Caribbean archaeology is riding the wave of an exponential curve, as myriad Caribbeanists had begun to address questions of adaptation, evolution, social and political organization, mythology, cosmology, and ideology (Ayubi and Haviser 1991; Robinson 1991; Siegel 1989). No longer is research in the Caribbean based almost exclusively on conventional survey and reconnaissance methods such as trial trenching, shovel test pitting, field walking, and ground surveys. Indeed, the various papers presented at the Twenty-First Congress of the International Association for Caribbean Archaeology (IACA), held in Trinidad and Tobago in July 2005, underscored the extent to which research agendas are being increasingly informed by a holistic mix of archaeological data, field methods, and scientific techniques, including geoinformatics.

The following examples favorably reflect the growing popularity of geoinformatics within the context of Caribbean archaeology. In 1998, geophysical surveys of a Jewish cemetery in Nevis resulted in the identification of at least 44 possible burials within the cemetery in addition to the 19 marked burials (Terrell 1998). Recent studies of St. Kitts’s prehistoric settlement patterns incorporated not only reconnaissance site visits and field surveys but also the use of GIS-generated maps, including the production of 500-m buffer zones depicting possible site catchment zones (Farag and Ramlal 2005). Another example pertains to the innovative creation of a system by Landon and Seales (2005) for building three-dimensional models for Caribbean petroglyphs based on reconstructions of Taino petroglyphs at Caquana, Puerto Rico. According to Landon and Seales (2005) these digital models were aimed at allowing digital access to and preservation of petroglyphs in remote areas that often remain unprotected from the elements. By coupling aerial photographs with highly accurate survey techniques, large-scale area excavation, and a fully automated barcode-based computer system, Kappers, Fitzpatrick, and Kaye (2007) successfully created a three-dimensional model of the fast-disappearing site of Grand Bay in Carriacou. Essentially, the resulting GIS data set provided the means to construct three-dimensional modeling of the site, a necessary component for developing future strategies dedicated to investigating and protecting archaeological sites on the island (Kappers et al. 2007). Given these important developments, this volume is both timely and relevant, as it epitomizes a significant trend in Caribbean archaeology that is proving to be increasingly useful to scholarly pursuits in the region.

Another important reason for this volume relates to the negative impacts on archaeological sites of sprawling urban growth, agriculture, mining, and land erosion in various Caribbean territories. These impacts have been particularly damaging to pre-Columbian sites, as these sites tend to be generally less visible on the landscape than their historic period counterparts. For instance, since the first systematic archaeology was conducted on the island in 1985, Anguilla has experienced a dramatic loss of its Amerindian archaeological heritage as a result of two decades of hotel development along the coastline where the sites are situated (Crock 2005). Michiel Kappers (2004) indicated that in March/April 2003, 11 site locations in Carriacou were surveyed and mapped. Most sites in Carriacou were found to be endangered by erosion through wave, storm, and tidal action and sand dredging by the local population, with the site of Grand Bay being the primary example.

The realization that archaeological resources are finite and diminishing (Drewett 2001:7) has prompted greater cultural resource management efforts in some Caribbean territories, as evidenced by the creation of heritage and planning organizations such as the Jamaica National Heritage Trust, the Institute of Jamaica, and the Town and Country Planning Authority in Jamaica; the National Trust, the Town and Planning Division, and the Tobago Trust in Trinidad and Tobago; the Barbados Museum and Historical Society and the Archaeological Museum of Aruba. There is also growing recognition among archaeologists, heritage managers, and policy makers in the Caribbean that geoinformatics is the preferred technique in heritage management, as it generally allows for more efficient data collection, analysis, and retrieval than conventional paper-based methods. Aerial photography is particularly useful for preventive prospecting, which means detecting and documenting sites before their destruction. These may be especially relevant in the face of the growing destruction of archaeological sites as a result of flooding, land erosion, mining, and road and building construction, which often result in the complete destruction of archaeological sites, despite the presence of heritage management organizations throughout the region.

Caribbean geography is also well suited to the application of geoinformatics. The island chain is divided into three parts. The Lesser Antilles form an arc extending northward from Trinidad and Tobago to the Virgin Islands; the Greater Antilles stretch westward from there to the Yucatan Peninsula; and the islands of the Bahamian Archipelago form a triangle in the Atlantic Ocean between the Greater Antilles and Florida (Rouse 1992:3). Although the islands in the Caribbean are generally small, their myriad microenvironments such as river valleys, forested areas, grasslands, coastlines, plains, hills, and mountains often pose significant challenges with respect to both site visibility and accessibility. Visibility refers to the extent to which a site has been buried or covered by soil aggradation and vegetation since its occupation (MacManamon 1984:224). James Zeidler (1995:11) discusses how low to nonexistent visibility because of dense vegetation cover is a common problem throughout the neotropics, which includes much of the Caribbean. This can be caused by old-growth forest vegetation with dense understory, more recent secondary-growth forest, or the dense ground cover of tall grasses used for pasturage. Trinidad and Tobago, which was connected to the South American mainland as recently as 10,000 years ago and remains in close proximity to the South American mainland, can be cited as a case in point. Comprising approximately 90 percent of the land area (Davis et al. 1986), the forests of the twin island republic not only share the same flora with South America but also, more significantly, these forests conceal many archaeological sites on the ground. Often, even cultivated plots exhibit completely obscured ground surfaces (Zeidler 1995:11). For instance, extensive sugar and rice cultivation on the plains of St. Catherine in south-central Jamaica has significantly reduced the visibility of archaeological sites within this particular area.

Accessibility, on the other hand, can be defined as the ability to physically inspect a given area of terrain (Zeidler 1995:12). Cases of low accessibility can be caused by difficult terrain or dense vegetational growth, which may impede or reduce mobility. Recent landscape modifications in the Caribbean may have completely destroyed evidence of archaeological occupations or covered them entirely with large expanses of soil, water, or modern construction. Because of ongoing land erosion, much of Canoe Bay, a large Amerindian site on the southwest coast of