The Pennsylvania State University

The Graduate School

Intercollege Graduate Program in Ecology

NATIVE AMERICAN LAND USE LEGACIES IN THE PRESENT DAY LANDSCAPE

OF THE EASTERN UNITED STATES

A Dissertation in

Ecology

by

Sarah E. Johnson

 2014 Sarah E. Johnson

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

May 2014

The dissertation of Sarah E. Johnson was reviewed and approved* by the following:

Marc D. Abrams Steimer Professor of Forest Ecology Dissertation Advisor Chair of Committee

Margot Kaye Associate Professor of Forest Ecology

Dean Snow Professor Emeritus of Anthropology

Kim Steiner Professor of Forest Biology

David M. Eissenstat Professor of Horticulture Chair, Intercollege Graduate Program in Ecology

*Signatures are on file in the Graduate School

iii ABSTRACT

Native American populations in the eastern United States were active and passive land managers during the period from AD 500 to 1650. The impacts of these land uses were profound, and identifiable in historic records. In this study, vegetation and soil characteristics from archaeological sites and off-site areas were quantified on three United States Department of

Defense installations and one protected archaeological site in the eastern U.S. to determine if this legacy of Native American land uses exists on present day landscapes.

Vegetative indicator species were identified through ethnobotanical literature review to determine which species were highly important in the diet and life of Native American groups of interest. This research defined two major vegetation indicators, white oak (Quercus alba L.) at

MCB Quantico, Virginia and the oak-pine-blueberry (Quercus-Pinus-Vaccinium) association of a sandy glacial outwash moraine at Fort Drum, New York. At both of these installations, forested archaeological sites were shown to be distinctly more open (decreased trees per acre) than off-site plots. Soil charcoal, while showing higher incidence on archaeological sites than on off-site areas, was fairly abundant across all surveyed areas, possibly indicating the influence of widespread Native American burning.

This research also suggests that Native Americans may have had an influence on soil characteristics in areas with intensive habitation and inherently low fertility soils. At Fort Drum archaeological sites presently have soils that display darker color and higher over-all fertility.

Native American influences were less discernible at study sites that exist along waterways. This may be due to their inherently more fertile soils, which would not require amendments to facilitate agriculture, or increased importance of fresh or brackish water resources, decreasing the need for intensive agriculture.

iv Present-day vegetation and soils associated with Late Woodland and Mississippian archaeological sites show different overall trends, influenced by settlement patterns, available resources, and baseline soil fertility. Application of these methodologies therefore requires careful examination of geographic variables and anthropological context. Predictive modeling techniques currently used by cultural resources managers could be improved through addition of other important landscape variables, such as vegetation and soils. This research provides a more complete understanding of Native American land use legacies on the present day landscape in the vegetation and soils, and highlights important indicator species and soil fertility trends at different locations in the eastern United States.

v TABLE OF CONTENTS

LIST OF FIGURES ...... vii

LIST OF TABLES ...... x

ACKNOWLEDGEMENTS ...... xiii

Chapter 1 Introduction ...... 1

Native American land use legacies in the eastern United States ...... 1 Summary of Chapters ...... 4 Literature Cited ...... 7

Chapter 2 Description of Study Areas and Witness Trees Analysis ...... 12

Fort Drum Army Installation, New York ...... 15 Witness Tree Analysis ...... 22 Canfield’s Island, Lycoming County, Pennsylvania ...... 27 Marine Corps Base Quantico, Virginia ...... 31 Witness Tree Analysis ...... 36 Cheatham Annex Naval Supply Station, Virginia ...... 39 Witness Tree Analysis ...... 46 Summary ...... 50 Literature Cited ...... 51

Chapter 3 The Relationship Between Vegetation, Soil Charcoal, and Archaeological Site Location ...... 59

Introduction ...... 59 Objectives ...... 63 Methods ...... 64 Results and Discussion ...... 73 Identification of Indicator Species ...... 73 Comparison of Archaeological and Off-site Overstory Composition ...... 76 Comparison of Archaeological and Off-site Soil Charcoal ...... 83 Classification Tree Modeling ...... 86 Conclusion ...... 93 Literature Cited ...... 95

Chapter 4 Native American Black Earth in the Eastern United States ...... 103

Introduction ...... 103 Objectives ...... 105 Methods ...... 106 Results and Discussion ...... 118 Comparison of Overall Soil Data from Archaeological and Off-site Areas...... 118

vi Regression Tree Model Results ...... 121 Spatial Interpolation Results ...... 125 Conclusion ...... 142 Literature Cited ...... 144

Chapter 5 Conclusion ...... 150

A protocol for efficient identification and protection of cultural resources ...... 150 Literature Cited ...... 158

Appendix: Ethnobotanical Inventory and Data Sources ...... 161

A. Ethnobotanical information gathered through personal communication ...... 161 New York ...... 161 Pennsylvania ...... 162 Virginia ...... 163 B. Literature utilized for ethnobotanical inventory ...... 164

vii

LIST OF FIGURES

Figure 2-1. Map of Fort Drum land ownership and training areas, including documented locations of prehistoric archaeological sites, and inset map indicating the location of Fort Drum in upstate New York (data on archaeological site locations supplied by the Fort Drum CRM office)...... 17

Figure 2-2. Oak, pine, and blueberry dominating the sandy soils of the glacial outwash moraine on Fort Drum, New York...... 18

Figure 2-3. Map of the various Tracts and land purchases in northern New York in the 1700s...... 23

Figure 2-4 Aerial view of the Canfield’s Island Archaeological Site, indicating the agricultural matrix the site currently lies in. The inset map shows the location of Lycoming County, Pennsylvania...... 28

Figure 2-5. Map of Marine Corps Base Quantico indicating the location of prehistoric archaeological sites. The inset shows the location of MCB Quantico along the Potomac River in Virginia (data on archaeological site locations supplied by the MCB Quantico CRM office)...... 32

Figure 2-6. Tuckahoe (utilized as important starchy food source) dominates the north side of Chopawamsic Creek on MCB Quantico, Virginia...... 34

Figure 2-7. Map of the Cheatham Annex Naval Supply Station, York County, Virginia. The Wilderness Area is noted with the turquoise border, and the inset notes the location of the Cheatham Annex within the state of Virginia (data on archaeological site locations supplied by Cheatham Annex CRM personnel)...... 40

Figure 2-8. American holly, loblolly pine, oak and beech on an archaeological site discovered through identification of shell midden evidence on Cheatham Annex, Virginia...... 42

Figure 2-9. Shell midden evidence is present in many places along the south side of Queen Creek on Cheatham Annex, Virginia...... 43

Figure 2-10. Derivative map of the rectified version of John Smith’s map of the Chesapeake Bay area and of the English settlement at Jamestown (original version by Smith 1612; rectified version produced by Virtual Jamestown 2000); indicating the location of Werowocomoco directly across the York River from the Cheatham Annex, where no settlements were found by Smith...... 46

viii Figure 3-1. Importance values, or mean decrease in accuracy of the model, for predictor variables in a classification tree model predicting Treatment (archaeological or off- site area) for data collected at Fort Drum...... 88

Figure 3-2. Importance values, or mean decrease in accuracy of the model, for predictor variables in a classification tree model predicting Treatment (archaeological or off- site area) for data collected at MCB Quantico...... 90

Figure 4-1. Munsell 10YR (red-yellow) system of assigning color to soil samples. On the left are color descriptions, and on the right are color plates for comparison to soil samples in the field...... 109

Figure 4-2. Location of soil samples on Camp Drum 1 Archaeological Site on Fort Drum Army Installation, NY...... 112

Figure 4-3. Location of soil samples on Canfield Island Archaeological Site in the Susquehannock River near Williamsport in Lycoming County, PA...... 113

Figure 4-4. Location of soil samples in an area of “hamlet” type archaeological sites along the north bank of Chopawamsic Creek on MCB Quantico, VA...... 114

Figure 4-5. Location of soil samples in the Wilderness Area on Cheatham Annex Naval Supply Station, VA...... 115

Figure 4-6. Spatial patterns in deviation from baseline values of pH, cation exchange capacity (meq/100 g), and soil color on Camp Drum 1 and surrounding area, Fort Drum, NY. Red areas indicate high values, while green areas indicate lower values. Colors of yellow to red indicate areas where measured soil characteristics were higher than baseline values. For soil color, more negative values indicate darker colors, so green, light green, and yellow are indicative of areas of soil darker than baseline soils...... 128

Figure 4-7. Spatial patterns in deviation from baseline values of calcium, magnesium, potassium, and phosphorus (ppm) on Camp Drum 1 and surrounding area, Fort Drum, NY. Red areas indicate high values, while green areas indicate lower values. Colors of yellow to red for calcium, potassium, and phosphorus indicate areas where measured soil characteristics were higher than baseline values. For magnesium, colors from green to dark orange indicate areas where magnesium values were lower than baseline, and only red areas indicate where magnesium measured was higher than baseline...... 129

Figure 4-8. Spatial patterns in pH, cation exchange capacity (meq/100 g), and soil color on a Susquehannock village site on Canfield Island, Lycoming County, PA. Blue areas indicate high values, while green areas indicate lower values. For soil color, lower values are concurrent with darker soil colors, so areas of green are indicative of darker soil. Baseline soil characteristics were not available in this area due to modern land disturbance surrounding the archaeological site...... 132

ix Figure 4-9. Spatial patterns in calcium, magnesium, potassium, phosphorus (all in ppm) on a Susquehannock village site on Canfield Island, Lycoming County, PA. Blue areas indicate high values, while green areas indicate lower values. Baseline soil characteristics were not available in this area due to modern land disturbance surrounding the archaeological site...... 133

Figure 4-10. Spatial patterns in deviation from baseline values of pH, cation exchange capacity (meq/100 g), and soil color on the site of several hamlet-style habitations on the north side of Chopawamsic Creek, Quantico, VA. Red areas indicate higher values, while green areas indicate lower values. Colors of yellow, orange, and red indicate areas where measured soil characteristics displayed values higher than baseline characteristics. For soil color, lower values indicate darker colors, and negative values indicate areas where soil was darker than baseline; therefore areas of yellow to green indicate darker color ...... 136

Figure 4-11. Spatial patterns in deviation from baseline values of calcium, magnesium, potassium, and phosphorus on the site of several hamlet-style habitations on the north side of Chopawamsic Creek, Quantico, VA. Red areas indicate higher values, while green areas indicate lower values. Colors of yellow, orange, and red indicate areas where measured soil characteristics displayed values higher than baseline characteristics...... 137

Figure 4-12. Spatial patterns of pH, cation exchange capacity (meq/100 g), and soil color on the south bank of Queen Creek in the Wilderness Area, Cheatham Annex, VA. Larger symbols indicate samples with higher measured values of soil parameters of interest. For soil color, smaller values indicate darker colors, therefore smaller symbols are indicative of areas of darker soils...... 139

Figure 4-13. Spatial patterns of calcium, magnesium, potassium, and phosphorus (all in ppm) on the south bank of Queen Creek in the Wilderness Area, Cheatham Annex, VA. Larger symbols indicate samples with higher measured values of soil parameters of interest...... 140

Figure 5-1. EPA Level III Ecoregions of the eastern United States...... 151

Figure 5-2. Required inputs for development and use of a CRM Decision Support Tool. Boxes at the top highlight information required for fully developing the DST, and boxes at the bottom include information that would be required from a user of the DST...... 156

x

LIST OF TABLES

Table 2-1. Species, counts, and percentage composition in the presettlement forest for Great Tract No. 4 in northern New York, as surveyed in 1795...... 25

Table 2-2. Witness tree composition analysis for parcels surveyed within the sandy glacial moraine...... 27

Table 2-3. Witness tree counts for the area of Marine Corps Base Quantico from land patent and grant records from 1648 – 1700...... 38

Table 2-4. Species, counts, and percentage composition in the early Colonial Period in the Tidewater Area of Virginia, specifically York County, from land patents and grants from 1623 to 1700...... 49

Table 3-1. Physical characteristics of archaeological and off-site areas at Fort Drum, New York. Soil characteristics of FDP 1242 and 1244 were not available through Natural Resources Conservation Service spatial data, thus textures were determined from the soil samples taken at the sites themselves (these textures are indicated with ‘*’). Sites where a suitable off-site analog for comparison were not able to be found are indicated by a ‘-‘...... 67

Table 3-2. Physical characteristics of archaeological and off-site areas at MCB Quantico, Virginia...... 67

Table 3-3. Physical characteristics of sites discovered through traditional testing and through shell midden evidence on the Cheatham Annex, Virginia. Natural Resources Conservation Service spatial soils data for York County, VA is lacking information on the Annex. Therefore, soil textures of all samples were determined from samples taken in the field (Soil Texture column attributed with a ‘*’)...... 68

Table 3-4. Fort Drum Army Installation, NY indicator species and uses...... 74

Table 3-5. Marine Corps Base Quantico, VA indicator species and uses...... 75

Table 3-6. Cheatham Annex Naval Supply Station, VA indicator species and uses...... 76

Table 3-7. Compiled overstory data for archaeological sites (“Arch.”) and off-site areas (“Off-Site”) at Fort Drum, NY. First columns under each heading indicate raw data, second columns indicate relative data. Significant differences between importance values on archaeological sites versus off-site areas at an alpha level of 0.05 (p < 0.05) are indicated by [**] next to the species name, while significant differences at an alpha level of 0.1 (p < 0.1) are indicated by a [*]. n = 50 archaeological site plots and 25 off-site plots ...... 78

xi Table 3-8. Compiled overstory data for archaeological sites (“Arch.”) and off-site areas (“Off-Site”) at MCB Quantico, VA. First columns under each heading indicate raw data, second columns indicate relative data. Significant differences between importance values on archaeological sites versus off-site areas at an alpha level of 0.05 (p < 0.05) are indicated by [**] next to the species name, while significant differences at an alpha level of 0.1 (p < 0.1) are indicated by a [*]. n = 28 archaeological site plots and 20 off-site plots ...... 80

Table 3-9. Compiled overstory data for archaeological sites (“Arch.”) and sites identified through shell midden evidence (“Midden”) at Cheatham Annex Naval Supply Station, VA. First columns under each heading indicate raw data, second columns indicate relative data. Significant differences between importance values on archaeological sites discovered through traditional testing versus those discovered with shell midden evidence at an alpha level of 0.05 (p < 0.05) are indicated by [**] next to the species name, while significant differences at an alpha level of 0.1 (p < 0.1) are indicated by a [*]. n = 20 traditional archaeological site plots and 23 shell midden site plots ...... 82

Table 3-10. Soil charcoal results for archaeological and off-site areas on Fort Drum, NY. Samples were considered positive for soil charcoal if they contained coarse soil charcoal (>1 mm in diameter), and sites were considered positive if at least 50% of samples contained coarse soil charcoal. A ' - ' indicates that no suitable off-site area was located for comparison with the cultural site, due to military activities and/or construction...... 84

Table 3-11. Soil charcoal results for archaeological and off-site areas on MCB Quantico, VA. Samples were considered positive for soil charcoal if they contained coarse soil charcoal (>1 mm in diameter), and sites were considered positive if at least 50% of samples contained coarse soil charcoal. A ' - ' indicates that no suitable off-site area was located for comparison with the cultural site, due to military activities and/or construction...... 85

Table 3-12. Soil charcoal results for archaeological sites discovered through traditional testing and through identification of shell midden evidence on Cheatham Annex, VA. Samples were considered positive for soil charcoal if they contained coarse soil charcoal (>1 mm in diameter), and sites were considered positive if at least 50% of samples contained coarse soil charcoal. A ' - ' indicates that no suitable off-site area was located for comparison with the cultural site, due to military activities and/or construction...... 86

Table 4-1. Munsell 10YR soil color code and corresponding analysis code for statistical analysis...... 109

Table 4-2. Mean values (+/- standard error) of soil parameters of interest and p-values associated with t-tests for significant difference in means between for cultural sites and associated off-site areas at Fort Drum, NY. Soil sample depth is 20 centimeters into the A horizon. n = 32 archaeological samples and 30 off-site samples...... 119

xii Table 4-3. Mean values (+/- standard error) of soil parameters of interest and p-values associated with t-tests for significant difference in means between for cultural sites and associated off-site areas at MCB Quantico, VA. Soil sample depth is 30 centimeters into the A horizon. n = 28 archaeological samples and 20 off-site samples...... 120

Table 4-4. Mean values (+/- standard error) of soil parameters of interest and p-values associated with t-tests for significant difference in means between off-site areas at Fort Drum, NY and MCB Quantico, VA...... 121

Table 4-5. Percent increase in mean square error (MSE) of all predictors for each response variable (soil characteristic), lumping data from both Fort Drum, NY and Quantico, VA. Predictors of highest importance for each response variable are indicated by bold red text...... 122

Table 4-6. Percent increase in mean square error (MSE) of all predictors for each response variable (soil characteristic) for archaeological and off-site data from Fort Drum, NY. Predictors of highest importance for each response variable are indicated by bold red text...... 124

Table 4-7. Percent increase in mean square error (MSE) of all predictors for each response variable (soil characteristic) for archaeological and off-site data from MCB Quantico, VA. Predictors of highest importance for each response variable are indicated by bold red text...... 125

Table 4-8. Mean values (+/- standard error) and ranges of values of soil parameters of interest for fertility comparisons between grids of soil samples taken at significant archaeological sites at Fort Drum, NY, Canfield Island, PA, MCB Quantico, VA, and the Cheatham Annex, VA...... 142

Table 5-1. Essential factors of the research protocol employed in this study, categorized by the data collection method. To initially assess indicator vegetation and soil fertility characteristics of archaeological sites, data from all three columns should be collected. For CRM personnel wishing to employ a Decision Support Tool utilizing this information, data collection would be limited to column three, fieldwork...... 155

xiii

ACKNOWLEDGEMENTS

This project would not have been possible without the support, advice, and encouragement of my advisor Marc Abrams, and my committee members Margot Kaye, Dean

Snow, and Kim Steiner. Laurie Rush, Meg Schulz, and others with Fort Drum Archaeological

Division in New York provided information and staff time, as well as performed archaeological field surveys to identify cultural resources at several sites prioritized through the methods developed in this work. Staff at the Thomas T. Taber Museum of the Lycoming County

Historical Society, Paul Glunk, and Noel Strattan at the Pennsylvania Bureau for Historic

Preservation provided information and access to the Pennsylvania Cultural Resources GIS system. John Haynes and MCB Quantico Archaeological staff, Bruce Larson, Pam Anderson, and Dean Wright at Naval Weapons Station Yorktown, as well as Rhonda Mickelborough at the

Cheatham Annex Naval Supply Station provided information and staff time in support of my work at the two field sites in Virginia. Laura Leites and Patrick Drohan provided statistical and research advice. Funding was provided by the Department of Defense Legacy Program, and my time at the Pennsylvania State University was supported through a College of Agricultural

Sciences teaching and research assistantship. I would also like to thank the Penn State Ecology

Program, the program chair David Eissenstat, as well as Jean Pierce.

Throughout my time in graduate school at Penn State, my sanity was maintained through the love and support of my family, friends, my significant other Eric, and my dog Deme. I would also like to thank our nation’s active service men and women, and veterans.

Chapter 1

Introduction

Native American land-use legacies in vegetation and soils of the

eastern United States

Native Americans in the eastern United States were active and passive land managers in the pre-European settlement landscape, and the legacy of these uses can still be seen in the present day (Day 1953, Chapman et al. 1982, Delcourt 1987, Patterson and Sassaman 1988,

Denevan 1992, Fuller et al. 1998, McLauchlan 2003, Black et al. 2006, Bean and Sanderson

2008). They drew upon the forest for a great variety of products, and were capable of modifying environmental conditions (in some cases in spite of climate influence), dispersing plant species to new areas, and creating evolutionary changes in flora through artificial human selection (Smith

1989, Smith 2006, Abrams and Nowacki 2008). Forest management including the transplanting of valuable species, frequent burning for many reasons, clearing and fuelwood cutting near areas of habitation, and agricultural activity by Native Americans may have facilitated the establishment of early successional tree species including important dietary mast and fruit trees in the present-day eastern United States (Maxwell 1910, Peterson 1991, Black and Abrams 2001,

Williams 2002). Mast-bearing tree species may have been cultivated or favored by Native

Americans, increasing their importance in present-day forests (Munson 1986, Loeb 1998). In addition, Native Americans used fire to improve browse for game, to encourage growth of berries, mast, and possibly pine species (Pinus spp L.), and to clear underbrush to maintain agricultural fields or facilitate hunting (Clark and Royall 1995, Delcourt and Delcourt 1998,

Brown 2000, Williams 2002). These activities resulted in significant alteration of vegetation composition and structure in intensively occupied areas during the Late Woodland (beginning

AD 500) and Mississippian (AD 1000 to 1650) cultural periods until the time of European

2 contact. The legacy of these intensive land uses is still visible on the landscape today (Black and

Abrams 2001, Black et al. 2006, Foster et al. 2003). From this knowledge combined with ethnobotanical research comes the idea that there are certain species of plants, known as indicator species, which would be present today in areas of significant Native American activity.

Indicator species can be identified through knowledge of the nature of Native American land uses, and how these uses would have affected species composition. The legacy of Native

American land use in the vegetation of the pre-European settlement landscape was created through direct and indirect means. For example, frequent understory burning directly created desirable growing conditions for several important early and mid-successional tree species such as oaks (Quercus spp L.), hickories (Carya spp Nutt.), American chestnut (Castenea dentata

Marsh. Borkh.), and pines (Dorney and Dorney 1989, Abrams 1992). Early cultivation methods and transplanting increased the occurrence of species such as chenopodium (Chenopodium spp

L.), paw paw (Asimina triloba L., Dunal), black walnut (Juglans nigra L.), and others (Munson

1986, Peterson 1991, Loeb 1998, Smith 2006). Indirectly, the clearing of villages and fields created open habitats suitable for establishment of early successional species (Black and Abrams

2001), and caching and storage of berries created concentrations of seeds. Purposeful propagation of important species, such as mast-bearing oaks and hickories, centered on what

Native American groups ate and used in their everyday lives. Thus, these are the plants that would have increased in importance in the pre-settlement and potentially the modern day landscape.

Similar to the creation of Native American land use signals in vegetation, a signal of

Native American activities may be present in the soils of the eastern United States. It is well documented that Native Americans of the Eastern Woodlands were active cultivators of the

“three sisters”, neotropical maize (Zea mays L.), beans (Phaseolus vulgaris L.), and squash

(varieties of Cucurbita pepo L.), as well as native cultigens of several species (Eastern

3 Agricultural Complex; Chomko and Crawford 1978, Smith 1984, Lewandowski 1987, Smith

1989, Zeder et al. 2006). In many areas of the world, intensive native agricultural practices included the addition of soil amendments such as charcoal, organic refuse, and human and animal waste (Mann 2000, Lehmann et al. 2003, Hecht 2004, Steiner et al. 2007). These amendments served to create soils that have been termed “anthrosols” because of the role of humans in their creation, the most notable of which are the “Black Earth” soils of the Amazon Basin (Lehmann et al. 2004). Black Earth soils are characterized by a deep, dark surface horizon, and higher organic matter and nutrient-holding capacity than typical tropical soils (Lehmann et al. 2003, Hecht

2004). Clearing forests and creating fields, despite the usage of raised bed technology, was a very difficult and energy-consuming process. For this reason, Native Americans of the Eastern

Woodlands region may have used similar soil amendments to allow the fields they created to remain productive for longer time periods. Also, in areas such as pre-contact tidewater Virginia,

Native American populations were quite high (Smith 1624, Mook 1944), making it difficult to find new areas to clear for the creation of fields every five years (time period postulated for field productivity in a swidden system; Sykes 1980). These factors most likely led Native Americans to amend soils to increase productive life of the fields, possibly creating a legacy of increased fertility.

Cultural resources management (CRM) personnel are employed by all United States

Department of Defense land holdings. They are charged with identifying and protecting all the ancestral places and historic archeological sites potentially eligible for the National Register of

Historic Places (NRHP) that occur on military land. This mission is carried out in conjunction with new construction projects, training maneuvers, and base operations, which makes efficiency a priority in the identification of these important cultural sites. Thus, CRM personnel are always looking for new and better ways to efficiently identify cultural sites and significant cultural resource and heritage areas. Currently, many methods are used to locate sites efficiently; GIS

4 technology has been utilized in this endeavor, as well as remotely sensed data such as satellite imagery, aerial imagery and lidar (see Kohler and Parker 1986, Zeidler 2001, Masini and

Lasaponara 2013, Comer and Harrower 2013). Indicator species can also be used as a tool for natural and cultural resources managers to better protect our heritage. Identification of indicator species would require simple vegetation surveys, rather than time-consuming and costly acquisition and evaluation of remotely-sensed data, and application of GIS models. The work contained in this dissertation furthers our understanding of the relationship between indicator species distributions and actual archeological site locations and attributes. This approach has the potential for improving the efficiency of archaeological survey strategies, streamlining the cultural site location process, and strengthening cultural site evaluation methods currently in place.

Summary of chapters

The time period of interest in this dissertation is AD 500 to the time of contact with

European settlers in the eastern United States (initial contact year varies in eastern North America depending on location; generally between 1550 and 1650). This range encompasses two different cultural types, the Late Woodland and Mississippian. This was a time of both cultivation and hunting and gathering for most eastern Native American groups. Four sites within the Eastern

Woodland environment have been selected to pursue this study. These are Fort Drum Army

Installation in upstate New York, Canfield’s Island in the Susquehanna River in Lycoming

County, Pennsylvania, Marine Corps Base Quantico on the Potomac River in Virginia, and the

Cheatham Annex Naval Supply Station in York County, Virginia. The three military installations have the advantage of detailed cultural resource inventory data through Cultural Resources

Management Programs mandated on all government-owned lands. Chapter two of this document provides detailed information on each of these areas concerning climate and ecoregion,

5 vegetation, soils, Native American groups and lifeways, relevant post-European settlement information, and witness tree analysis to identify pre-settlement forest composition (Lutz 1930,

Bourdo 1956). The objective of this chapter is to fully characterize each setting in which this research was conducted, to set the stage for relevant conclusions to be made.

The third chapter of this dissertation describes testing of the relationship between current vegetation and soil charcoal presence, and archaeological site location. The goal of this chapter is to elucidate the effects of Native Americans on the vegetation of the present day landscape by comparison of archaeological sites and paired off-site areas. Particular attention is paid to the distribution of indicator species. Quantifiable patterns in vegetation and soil charcoal on and off archaeological sites are the focus, but qualitative trends are identified as well. Overall, the hypothesis is that vegetation on archaeological sites was affected by Native American land uses such that the signal of these land uses persists to the present day. One of these land uses in particular, frequent understory burning, was characteristic of the majority of Native American populations (Delcourt and Delcourt 1997, Williams 2000), and the presence of soil charcoal at the depth of Native American occupation can signal a significant history of this land use type

(Fesenmyer and Christensen 2010).

Another over-arching objective of this research is to determine if a signal of Native

American intensive use and amendment exists in modern-day soils on archaeological sites.

Therefore, the fourth chapter of this dissertation deals with the identification of Native American influences in the soil through the creation of “Black Earth” in Native American agricultural fields and midden areas. Various types of cultural sites and methods are used to discern patterns and relationships in soil nutrient levels on and off different types of archaeological sites from villages to temporary camps. It is hypothesized that in areas of intensive use, soil amendments both direct

(applied purposefully to extend the productive life of agricultural fields) and indirect (middens

6 containing organic wastes) have created a signal of higher fertility in soils that has persisted to the present day.

Chapter five provides conclusions regarding the applicability of this work to Cultural

Resources Management objectives on military installations throughout the eastern United States.

The Appendix of this document is an ethnobotanical inventory that lists plant species useful to

Native Americans for food, medicinal, and material sources, catalogued by geographic region

(northern New York, Pennsylvania, and coastal Virginia). This research furthers the scientific community’s understanding of Native American land use legacies on the present day landscape, knowledge that is necessary to fully and accurately characterize human influence on natural resources. Without full knowledge of the effects of past land use, we cannot make a comprehensive plan for future management of natural and cultural resources.

7 Literature Cited

Abrams, M.D. and Nowacki, G.J. 2008. Native Americans as active and passive promoters of

mast and fruit trees in the eastern USA. The Holocene 18: 1123-1137.

Abrams, M.D. 1992. Fire and the development of oak forests. BioScience 42: 346-353.

Bean, W.T. and Sanderson, E.W. 2008. Using a spatially explicit ecological model to test

scenarios of fire use by Native Americans: An example from the Harlem Plains, New

York, NY. Ecological Modelling 211: 301-308.

Black, B.A. and Abrams, M.D. 2001. Influences of Native Americans and surveyor biases on

metes and bounds witness-tree distribution. Ecology 82: 2574-2586.

Black, B.A., Ruffner, C.M. and Abrams, M.D. 2006. Native American influences on the forest

composition of the Allegheny Plateau, northwest Pennsylvania. Canadian Journal of

Forest Research 36: 1266-1275.

Bourdo, E.A., Jr. 1956. A review of the General Land Office Survey and of its use in quantitative

studies of former forests. Ecology 37: 754-768.

Brown, H. 2000. Wildland burning by American Indians in Virginia. Fire Management Today 60:

29-39.

Chapman, J., Delcourt, P.A., Cridlebaugh, P.A., Shea, A.B. and Delcourt, H.R. 1982. Man-land

interaction: 10,000 years of American Indian impact on native ecosystems in the lower

Little Tennessee River valley, eastern Tennessee. Southeastern Archaeology 1: 115-121.

Chomko, S.A. and Crawford, G.W. 1978. Plant husbandry in prehistoric eastern North America:

New evidence for its development. American Antiquity 43: 405-408.

Clark, J.S. and Royall, P.D. 1995. Transformation of a northern hardwood forest by aboriginal

(Iroquois) fire: Charcoal evidence from Crawford Lake, Ontario, Canada. The Holocene

5: 1-9.

8 Comer, D.C. and Harrower, M.J. 2013. Mapping Archaeological Landscapes from Space.

Springer, New York, New York.

Day, G.M. 1953. The Indian as an ecological factor in the northeastern forest. Ecology 34: 329-

346.

DeBoer, W.R. 1988. Subterranean storage and the organization of surplus: The view from eastern

North America. Southeastern Archaeology 7: 1-20.

Delcourt, H.R. and Delcourt, P.A. 1997. Pre-Columbian Native American use of fire on southern

Appalachian landscapes. Conservation Biology 11: 1010-1014.

Delcourt, P.A. and Delcourt, H.R. 1998. The influence of prehistoric human-set fires on oak-

chestnut forests in the southern Appalachians. Castanea 63: 337-345.

Denevan, W.M. 1992. The Pristine Myth: The landscape of the Americas in 1492. Annals of the

Association of American Geographers 82: 369-385.

Dorney, C.H. and Dorney, J.R. 1989. An unusual oak savanna in northeastern Wisconsin: The

effect of Indian-caused fire. American Midland Naturalist 122: 103-113.

Fesenmyer, K.A. and Christensen, N.L., Jr. 2010. Reconstructing Holocene fire history in a

southern Appalachian forest using soil charcoal. Ecology 91: 662-670.

Foster, D., Swanson, F., Aber, J., Burke, I., Brokaw, N., Tilman, D. and Knapp, A. 2003. The

importance of land-use legacies to ecology and conservation. BioScience 53: 77-88.

Fuller, J.L., Foster, D.R., McLachlan, J.S. and Drake, N. 1998. Impact of human activity on

regional forest composition and dynamics in central New England. Ecosystems 1: 76-95.

Hart, J.P. 1995. Storage and Monongahela subsistence-settlement change. Archaeology of

Eastern North America 23: 41-55.

9 Hecht, S.B. 2004. Indigenous soil management and the creation of Amazonian Dark Earths:

Implications of Kayapó practices. In Amazonian Dark Earths – Origin, Properties,

Management. Lehmann, J., Kern, D.C., Glaser, B. and Woods, W.I., editors. Kluwer

Academic Publishers, New York, New York. Pages 355-372.

Kohler, T.A. and Parker, S.C. 1986. Predictive models for archaeological resource location.

Advances in Archaeological Method and Theory 9: 397-452.

Lehmann, J., Kern, D.C., Glaser, B. and Woods, W.I. 2004. Amazonian Dark Earths – Origin,

Properties, Management. Kluwer Academic Publishers, New York, New York.

Lehmann, J., Pereira da Silva Jr., J., Steiner, C., Nehls, T., Zech, W. and Glaser, B. 2003.

Nutrient availability and leaching in an archaeological Anthrosol and a Ferralsol of the

Central Amazon basin: Fertilizer, manure and charcoal amendments. Plant and Soil 249:

343-357.

Lewandowski, S. 1987. Diohe’ko, the Three Sisters in Seneca life: Implications for a native

agriculture in the Finger Lakes region of New York State. Agriculture and Human Values

4: 76-93.

Loeb, R.E. 1998. Evidence of prehistoric corn (Zea mays) and hickory (Carya spp.) planting in

New York City: Vegetation history of Hunter Island, Bronx County, New York. Journal

of the Torrey Botanical Society 125: 74-86.

Lutz, H.J. 1930. Original forest composition in northwestern Pennsylvania as indicated by early

land survey notes. Journal of Forestry 28: 1098-1103.

Mann, C.C. 2000. The good earth: Did people improve the Amazon Basin? Science 4: 788-790.

Masini, N. and Lasaponara, R. 2013. Airborne lidar in archaeology: Overview and a case study.

Computational Science and Its Applications, Lecture Notes in Computer Science 7972:

663-676.

10 Maxwell, H. 1910. The use and abuse of forests by the Virginia Indians. The William and Mary

Quarterly 19: 73-103.

McLauchlan, K. 2003. Plant cultivation and forest clearance by prehistoric North Americans:

Pollen evidence from Ft. Ancient, Ohio, USA. The Holocene 13: 557-566.

Mook, M.A. 1944. The aboriginal population of Tidewater Virginia. American Anthropologist 46:

193-208.

Munson, P.J. 1986. Hickory silviculture: A subsistence revolution in the prehistory of eastern

North America. In Emergent Horticultural Economies of Eastern Woodlands. Keegan,

W.F., editor. Southern Illinois University, Carbondale, Illinois. Pages 1-20.

Patterson, W.A. and Sassaman, K.E. 1988. Indian fires in the prehistory of New England. In

Holocene Human Ecology in Northeastern North America. Nicholas, G.P., editor.

Plenum Press, New York, New York. Pages 107-135.

Peterson, R.N. 1991. Pawpaw (Asimina). In Genetic Resources of Temperate Fruit and Nut

Crops. Moore, J.N. and Ballington, J.R., editors. Acta Horticulturae 290: 567-600.

Smith, B.D. 2006. Eastern North America as an independent center of plant domestication.

Proceedings of the National Academy of Sciences 103: 12223-12228.

Smith, B.D. 1989. Origins of agriculture in eastern North America. Science 246: 1566-1571.

Smith, B.D. 1984. Chenopodium as a prehistoric domesticate in eastern North America: Evidence

from Russell Cave, Alabama. Science 226: 165-167.

Smith, J. 1624. The Generall Historie of Virginia, New England, and the Summer Isles: together

with The true travels, adventures and observations, and A sea grammer. MacLehose,

Glasgow, Scotland.

11 Steiner, C., Teixeira, W.G., Lehmann, J., Nehls, T., Vasconcelos de Macêdo, J.L., Blum, W.E.H.

and Zech, W. 2007. Long term effects of manure, charcoal and mineral fertilization on

crop production and fertility on a highly weathered Central Amazonian upland soil. Plant

and Soil 291: 275-290.

Sykes, C.M. 1980. Swidden horticulture and Iroquoian settlement. Archaeology of Eastern North

America 8: 45-52.

Williams, G.W. 2002. Aboriginal use of fire: are there any ‘natural’ plant communities? In

Wilderness and political ecology: Aboriginal land management – myths and realities.

Kay, C.E. and Simmons, R.T., editors. University of Utah Press, Salt Lake City, Utah.

Pages 179-211.

Williams, G.W. 2000. Introduction to aboriginal fire use in North America. Fire Management

Today 60: 8-12.

Zeder, M.A., Emshwiller, E., Smith, B.D. and Bradley, D.G. 2006. Documenting domestication:

The intersection of genetics and archaeology. TRENDS in Genetics 22: 139-155.

Zeidler, J.A. 2001. Dynamic modeling of landscape evolution and archaeological site

distributions: A three-dimensional approach. SERDP Project CS-1130. Center for

Environmental Management of Military Lands TPS 01-8, Colorado State University, Fort

Collins, Colorado.

Chapter 2

Description of Study Areas and Witness Tree Analysis

Four sites in the eastern United States were chosen to conduct this research: Fort Drum

Army Installation, New York, Canfield’s Island Archaeological Site, Lycoming County,

Pennsylvania, Marine Corps Base Quantico, Virginia, and the Cheatham Annex Naval Supply

Station, York County, Virginia. The time period of interest is AD 500 to contact with European settlers, which occurred between about 1600 and 1700, depending on location. Two cultural types are in evidence at these sites, the Late Woodland and Mississippian. The onset of the

Mississippian was delayed in areas that were further from the core of the development of

Mississippian cultural traits such as maize-based agriculture, changes in tool usage, village nucleation, and pottery styles (Rieth 2002). In some outlying areas, the cultural traits of the Late

Woodland extend to the time of European Contact, although maize-based agriculture was eventually adopted.

The eastern United States is now accepted as one of several independent centers of plant domestication in the world (Smith 1989, Smith 2006, Zeder et al. 2006). The Eastern

Agricultural Complex of indigenous seed crops is composed of several native plants that were originally harvested from the wild, and after long periods of selective collection and cultivation, were modified for human purposes. These plants are squash (Cucurbita pepo L.), little barley

(Hordeum pusillum Nutt.), goosefoot or lambsquarters (Chenopodium berlandieri Moq.), erect knotweed (Polygonum erectum L.), maygrass (Phalaris caroliniana Walter), sumpweed or marsh elder (Iva annua L.), and sunflower (Helianthus annuus L.). Subtropical crops, maize (Zea mays

L.), beans (Phaseolus vulgaris L.), and squash (C. spp L., of a different subtropical variety), were imported from Mesoamerica between 700 and 1100 AD (varies depending on location;

Lewandowski 1987, Struever and Vickery 1973, Loeb 1998). The spread of maize-based

13 agriculture technology in eastern North America progressed from the Southeast to the Northeast.

This can partially be attributed to cooler temperatures and decreased growing season length in northern areas; however over time, maize varieties were developed that could take advantage of the shorter growing season and production increased to the level at which surplus was produced

(Noble 1975, Stothers and Yarnell 1977).

Many archaeological sites for studying the Eastern Agricultural Complex can be found in the eastern United States (Delcourt et al. 1998, Chapman et al. 1982, Gremillion et al. 2008). In fact the greatest understanding of the eastern agricultural complex and eastern indigenous seed cropping systems has been documented through work in cave dwellings of Kentucky and the

Ozark Mountains (Fritz 1984, Gremillion 2004). Grain storage areas in these caves were preserved by the cool dry conditions in the caves, allowing archaeologists to determine the composition of the Eastern Agricultural Complex through plant genomics (Barber and Hubbard

1997, Zeder et al. 2006). At the time of contact, maize-based agriculture was very important as a reliable food source for the natives of eastern Virginia and Maryland (Smith 1608). Maize was cultivated along with subtropical Mesoamerican varieties of climbing beans and squash, and tobacco (Nicotiana rustica L.) where growing conditions were suitable (Watson 1989, Riley et al.

1990, Smith and Crawford 1997, Hart and Scarry 1999).

In addition to cultivated crops, Native Americans utilized many tree species, and affected the structure and composition of forests (Black and Abrams 2001, Black et al. 2006). Witness tree analysis was conducted at study sites where vegetation sampling took place (Ft Drum,

Cheatham Annex, and MCB Quantico), to understand and provide context for analysis of current vegetation trends and comparison of current vegetation with prehistoric in areas of Native

American habitation and use. Witness trees (or bearing trees) recorded in early land surveyor’s notes are used to reconstruct presettlement forest composition, including those affected by Native

American impacts (Lutz 1930, Bourdo 1956, Abrams and Ruffner 1995). Witness trees provide a

14 representation of forest composition and structure, and can also reveal historical disturbance patterns and the nature and extent of Native American influences (Black and Abrams 2001,

Foster et al. 2004). Witness trees are a source of quantitative data on presettlement conditions where much or all original forests have been cleared. The type of witness tree data considered here is metes and bounds data, in which trees and other identifying features were used to mark the boundaries and corners of property. Other sources of information about presettlement forest composition and structure come from early settler accounts, and General Land Office data which was especially prevalent in the Midwest where frontier land was partitioned and sold quickly

(White and Mladenoff 1994, Delcourt and Delcourt 1996, Schulte and Mladenoff 2001). Studies using this information have indicated significant changes in forest composition as a result of

European land uses, through comparison of witness tree records with present-day forest composition (Foster 1992, Dyer 2001, Whitney and DeCant 2003, Leahy and Pregitzer 2003).

Because of the method used to survey land, these surveys were not an intentional sampling of forest composition and are neither random nor impartial (Bourdo 1956, Black and

Abrams 2001). Surveyors may have had a tendency to choose only overstory individuals or the largest individuals, or trees of only one species. Bias in which trees were chosen may also have existed based on vigor of trees, abundance, bark characteristics, economic value, or any personal preference (Russell 1981, Grimm 1984). They also may have improperly identified trees (Lutz

1930, Loeb 1987). Determining whether or not these types of bias existed can be difficult, although one can evaluate the relative number of species recorded in the surveys, and whether or not the species of trees recorded often reach the canopy or were used as materials sources.

Examples of trees included in surveys that would indicate relatively low levels of this type of bias are sassafras (Sassafras albidum Nutt., Nees), dogwood (Cornus spp L.), and myrtle (Myrtus spp

L.), because they are typically found in the midstory and were not considered to have economic value (Burns Honkala 1990). Another bias that was sometimes present is in variation in witness

15 tree densities (number of trees recorded per unit area) with topography, in colonial era (pre-1785) metes and bounds type surveys (Black and Abrams 2001). This bias occurred because of the difficulty of moving over certain types of terrain, or the tendency of early land surveys to be located in more desirable landscapes, such as richer river valleys versus side slopes or ridge tops.

These sources of bias have led some researchers to believe that the metes and bounds surveys conducted in areas toward the coast of New England and in the Mid-Atlantic (i.e., the earliest surveys conducted) do not lend themselves to quantitative analysis (Stewart Ware, The

College of William & Mary, personal communication). This type of survey system was used in the counties that the two sites in Virginia fall into, Marine Corps Base Quantico in Stafford,

Fauquier, and Prince William Counties, and the Cheatham Annex Naval Supply Station in York

County. Virginia's land was parceled up and cleared for agriculture long before there was agreement on the standard township and range procedure for land surveys, thus land survey data and land deeds (and consequently, witness tree data) is scattered and usually incomplete for a given area in coastal and tidewater Virginia. The surveys conducted in the large tracts of northern New York however, were conducted later and although the methods used in these surveys were similar to those used in the General Land Office surveys of the Midwest, standardization was still lacking, and maps often did not accompany the surveyor’s notes. The witness tree data obtained for each of these sites was therefore used to simply identify the general species composition of forests during the time period of contact between Native Americans and

European settlers.

Fort Drum Army Installation, New York

Fort Drum Army Installation is located in upstate New York, in Jefferson, Lewis, and St.

Lawrence Counties (Figure 2-1). Fort Drum lies at roughly 44 degrees north latitude and 75.8 degrees west longitude, and the climate in this area is temperate humid continental. Yearly

16 temperatures typically range from -11 to 27 degrees Celsius (11 to 80 degrees Fahrenheit), with an average of 7.2 degrees (45.02 degrees Fahrenheit), as calculated with data from 1980 to 2010.

This area receives an average of 94.7 centimeters (37.3 inches) of rain yearly, and 214.1 centimeters (84.3 inches) of snow, calculated with data from 1980 to 2010 (NOAA National

Climatic Data Center). During the last glacial maximum, which occurred around 22,000 years ago (Hill 2006), glaciers covered much of North America, including what is now Fort Drum in upstate New York (Late Wisconsin). A moraine was formed by the recession of the glacier, and runs from north to south, transecting the southern portion of the land that Fort Drum now occupies. To the west and northeast of the moraine the land becomes low-lying and swampy, and is dominated by hydric soils. To the east of the moraine are the foothills of the Adirondacks. The dominant soils of this area are either loamy and productively farmed, or clayey in the floodplains of the river valleys (Pearson and Cline 1961, McDowell 1988). The moraine itself is composed of glacial outwash, and the soil texture is very sandy and highly leached (McDowell 1988). The vegetation associations are typical of the northern hardwood region, with high importance of sugar maple (Acer saccharrum Marsh.) and beech (Fagus grandifolia Ehrh.). Hemlock (Tsuga canadensis L. Carrière), basswood (Tilia americana L.), and red maple (A. rubrum L.) dominate in low-lying areas and swamps, while on Fort Drum, oak (Quercus alba L., Q. rubra L., and Q. montana L.) and pine (Pinus strobus L., P. resinosa Sol. Aiton, and P. rigida Mill.) species dominate on sandy glacial outwash soils (Figure 2-2).

17

Figure 2-1. Map of Fort Drum land ownership and training areas, including documented locations of prehistoric archaeological sites, and inset map indicating the location of Fort Drum in upstate New York (data on archaeological site locations supplied by the Fort Drum CRM office).

18

Figure 2-2. Oak, pine, and blueberry dominating the sandy soils of the glacial outwash moraine on Fort Drum, New York.

This area of New York was inhabited by Native Americans of Iroquoian groups, who spoke many languages in the Northern Iroquoian family (Bamann et al. 1992). The Iroquoian language group encompassed several groups of people. After 1450 local groups in what is now

New York State banded together in alliances that eventually evolved into the Iroquois League or

Confederacy (Trigger 1978). Others, such as the St. Lawrence Iroquois, remained distinct. The

Iroquoia League or Confederacy included the Mohawk, Oneida, Onondaga, Cayuga, and Seneca tribes. The Tuscarora joined in 1722 after European settlement of coastal areas, but at that time

19 this area of present-day New York was still considered the frontier (Bamann et al. 1992). These groups were formerly hostile before forming the League of the Iroquois, but banded together due to the threat of European settlement. European diseases, such as smallpox, periodically spread among the Iroquoian groups, decimating the population. After the introduction of Old World diseases, the population of Native Americans in this area was reduced by up to 60-80% as early as the late 16th century (Snow and Lanphear 1988, Snow 1996). Native peoples aggregated into villages for safety and survival, many times irrespective of what tribe they had originated from

(Snow 1994).

Fort Drum itself is today considered ancestral ground of the St. Lawrence, Seneca, and

Oneida tribal nations (Ft. Drum Cultural Resources Management). These groups of Native

Americans tended to form semi-sedentary communities of up to about 200 individuals during the

6th to 10th centuries. The lifeways of these groups during this time remained typical of the Late

Woodland culture up until the time of European contact, although maize-based agriculture was practiced here. Hunting and gathering were the most important sources of food, with agriculture used to supplement food supply (Abel 2002). Wild game, such as deer (Odocoileus virginianus

Zimmermann), bear (Ursus americanus Pallas), eastern wild turkey (Meleagris gallopava silvestris Linnaeus), and small mammals, were favored game species. They were hunted frequently from semi-permanent satellite sites, and meat was dried for storage (Bamann et al.

1992, Abel 2000). Hunting was integral in the winter months because the supply of plant resources dwindled during this time. These people employed bow and arrow, and used group game drives and fire to increase harvest rates (Webster 1984-85).

These groups utilized many plant species for food, medicinal purposes, and building materials (Appendix). Species such as oaks, hickories (Carya spp Nutt.), and blueberry

(Vaccinium spp L.) were especially valued as food (mast or berries) and materials sources. Land use practices employed by the Iroquois, such as periodic burning, facilitated a supply of these

20 sources and had a marked effect on the landscape and composition of the forest (Day 1953, Pyne

1983, Black and Abrams 2001). These land use practices also ensured a continual supply of food that benefited populations of preferred game species such as deer and bear (Abrams and Nowacki

2008). After 700 AD, maize-based agriculture reached the Fort Drum area, and natives conducted raised-bed gardening, cultivating the “three sisters:” maize, beans, and squash (Trigger

1978, Bamann et al. 1992). Wild-gathered and cultivated food resources were stored in below- ground pits, and prepared using implements such as grinding stones and boiling stones (“hot rock” cooking) (Wandsnider 1997, Brumbach and Bender 2002, Gremillion 2004).

A portion of the present area of Fort Drum was purchased for use as a military training site in 1908, when it was named ‘Pine Camp’. It was so named because of the pine plains there: the sandy, glacial-outwash soils that supported white and pitch pine (P. strobus L. and P. rigida

Mill., respectively), as well as oaks. Further land was purchased in 1935, and then again at the outbreak of World War II (totaling 107,265 acres/43,409 hectares), when the name was changed to Camp Drum. Prior to Department of Defense acquisition of the land, wildfires impacted the landscape, as well as clearing of timber. Agriculture was limited because of the nature of the soils in the area, especially in the pine plains. However the surrounding area, especially to the west and south, was cropped. Homesteads sprung up in the area, and over 600,000 acres of land in the area was purchased by James LeRay de Chaumont in the late eighteenth century. LeRay's plan was to subdivide and sell the land to new immigrants, however much of it remained in forest. Modern agriculture has not occurred on Fort Drum (i.e., no large equipment, soil compaction, or inorganic fertilizers). The Civilian Conservation Corps planted several hundred acres of pitch pine on Fort Drum land in the year 1934; these areas were avoided in vegetation surveying. Military activities include surface maneuvers, construction, borrowing of material for construction fill and military activities, and logging, however once cultural resources management personnel were established (early 1990’s) historic and prehistoric archaeological

21 sites that were found have been protected from activities that would impact sub-surface structure and artifacts (Fort Drum Cultural Resources Management).

The Fort Drum Cultural Resources Program is charged with identifying and protecting all of the ancestral places and historic archeological sites potentially eligible for the National

Register of Historic Places that occur on Fort Drum. Cultural Resources Management (CRM) personnel have a goal of completing archaeological testing on the entire base, and currently have completed the majority of the area of Ft. Drum, outside of the central impact area. Traditional archaeological testing methods are employed by CRM personnel and field crews. Baseline transects are laid, and shovel test pits are dug at specified intervals along each transect. If any shovel pits return positive results (i.e., if material removed from pits contains artifacts or evidence of archaeological features), cruciforms are implemented. Cruciforms are a series of four pits at a one-meter distance from original shovel test pit, followed by a further series of pits five meters from the original pit. If cruciforms return positive results, more complete test units are used to determine the exact nature of resources present (Fort Drum Cultural Resources Management).

Currently, Fort Drum CRM personnel have identified more than 200 Native American ancestral places within the boundary of the installation. These include a ceremonial area known as the “Calendar Site”, and a St. Lawrence Iroquoian village, known as Camp Drum 1. The village was inhabited by about 200-500 people for about 15 years in the early 16th century. Many sites that have been located fall under the umbrella of “satellite sites” or what will be known as

“resource camps” throughout this document because of the nature of the activities that took place there. Examples are chert-gathering, building of boats or making of tools, and hunting and gathering. Many times these sites would have been located in areas where the particular resource of interest was abundant; however this wasn’t necessarily near any long-term habitation areas.

Therefore, temporary camps were erected in these areas and inhabited for days to months at a time, most likely on at least a yearly basis (Abel 2000, Rieth 2002). These camps can be

22 identified by artifacts such as lithic scatters from tool-making, hearths, and projectile points (Fort

Drum Cultural Resources Management). Because of the amount of archaeological testing that has been done on Fort Drum, areas that have tested negative for cultural resources have been located as well. For this reason, locating off-site areas negative for cultural resources, as data pairings for archaeological sites, was relatively straightforward.

Witness Tree Analysis

Colonists began to carve out small farms in this area of upstate New York, many times on abandoned Native American land, in the late-1600s to early 1700s although this settlement was on a relatively small scale when compared with coastal areas. Much of the unclaimed land in more remote areas was bought by speculators in large tracts, sometimes called Great Tracts

(Schneider 1997, Figure 2-3). Some areas were bought up in large chunks by the government to reward soldiers who served in the Revolutionary War (Sullivan and Magill 2013; http://www.rootsweb.ancestry.com/~nycayuga/land/mtracths.html). One such large tract was the

Military Tract in northern New York which was given to New York soldiers to repay them for service and to promote settlement. Witness tree analysis and presettlement forest composition research has been done using the original survey documents from this area, including the Holland

Land Company and the Phelps & Gorham Purchases of western New York (e.g., Seischab 1990,

Seischab and Orwig 1991, Marks et al. 1992).

23

Figure 2-3. Map of the various Tracts and land purchases in northern New York in the 1700s.

The land that Fort Drum now occupies was part of an original sale of about 3.6 million acres of land to Alexander Macomb, bought in 1791 and surveyed in 1795 (Figure 2-3). This purchase was broken into Great Tracts (1 through 6; Figure 2-3). Great Tract No. 4 encompasses all of the land that Ft. Drum currently occupies; therefore witness tree analysis was completed on the 1000 individual lots (of 440 acres each) that make up this Tract. Land surveyors were employed to record the boundaries of each parcel sold. The metes and bounds survey system was used, in which each individual lot was sectioned off from the whole in roughly the same shape and size, and markers were employed to locate boundaries and where corners met. These markers were trees, stakes, stakes in association with stone piles, and stone piles that would then be located based on distance and direction from a nearby tree. Surveyor field books recording the locations and boundaries of each lot in a purchase have been transferred onto microfilm and collected in the New York State Archives in Albany, NY. Each lot record includes survey notes and remarks on the land including dominant timber species present, name of the surveyor, name

24 and/or number of township the lot is in, tract or lot number, county, date survey was completed and filed, occasional rough maps of land, and occasional valuations of lots.

In Great Tract No. 4, there were 27 total species recorded, and 1192 total trees used as property boundary and corner markers (Table 2-1). All lot surveys used tree markers as corners, but many lots used multiple stakes as well. The variety of tree species used as markers indicates that bias on the part of surveyors towards particular tree species(s) is minimal (Black and Abrams

2001). All witness trees were compiled and percent composition of species in the presettlement forest was determined. Analysis of the witness tree data indicates that beech makes up the highest percentage of the presettlement forest at 27.8% of the trees surveyed. Important mast species such as oak and hickory are present, but much less important (highest importance is white oak at just 1.3% of the forest). In low-lying areas such as on the banks of the St. Lawrence River, species such as basswood, elm (Ulmus spp L.), hemlock, and maple dominate.

25

Table 2-1. Species, counts, and percentage composition in the presettlement forest for Great Tract

No. 4 in northern New York, as surveyed in 1795.

Species Count Percent Composition Beech (Fagus grandifolia Ehrh.) 331 27.8 Maple (Acer spp L.) 223 18.7 Hemlock (Tsuga canadensis L., Carrière) 150 12.6 Black ash (Fraxinus nigra Marsh.) 80 6.7 Birch (Betula spp L.) 63 5.3 Ironwood (Carpinus caroliniana Walter, or Ostrya virginiana Mill., K. Koch) 55 4.6 Basswood (Tilia americana L.) 49 4.1 Elm (Ulmus spp L.) 43 3.6 Pine (Pinus spp L.) 39 3.3 Ash spp (Fraxinus spp L.) 34 2.9 White pine (P. strobus L.) 27 2.3 Sugar maple (A. saccharum Marsh.) 21 1.8 White oak (Quercus alba L.) 16 1.3 Cedar (Thuja accidentalis L.) 11 0.9 Yellow birch (Betula alleghaniensis Britt.) 8 0.7 White ash (F. americana L.) 7 0.6 Poplar (Populus spp L.) 6 0.5 Hickory (Carya spp Nutt.) 5 0.4 Black oak (Q. velutina Lamb.) 4 0.3 Oak spp (Quercus spp L.) 4 0.3 Swamp oak (Q. bicolor Willd.) 4 0.3 Spruce (Picea spp Mill.) 3 0.3 Red maple (A. rubrum L.) 2 0.2 Red oak (Q. rubra L.) 2 0.2 Tamarack (Larix laricina Du Rol, K. Koch) 2 0.2 Silver maple (A. saccharinum L.) 2 0.2 Pitch pine (P. rigida Mill.) 1 0.1 Total 1192 100.0

This analysis shows that beech was the dominant species in most areas of this landscape in upstate New York, followed by other typical northern hardwood species such as maple, hemlock, ash (Fraxinus spp L.), and birch (Betula spp L.). This is significant because in this mainly northern hardwood dominated forest, the presence of a more early successional, non-

26 climax species, such as oak species, indicates possible Native American impacts and land uses that would perpetuate a disturbance-adapted forest type. Therefore, when catchments of species such as oak are encountered, as in the pine plains, significant Native American land uses may have impacted the landscape to create the anomaly.

The forest composition of much of this landscape is very different from that of a relatively small area that lies within the boundary of Fort Drum, the pine plains. Surveyed lots indicating poor conditions and sandy soils were catalogued separately from the above data. Only

6 lots recording these characteristics in the remarks on the lot were found. At 440 acres each

(178 hectares), this comprises 2,640 acres of land (1,068 hectares). Delineation of the extent of the glacial moraine using geographic data (based on soils data and field observations), indicates that the area of very poor soils that would have been remarked on by surveyors is roughly 4 square miles, slightly smaller than 6 lots of 440 acres (178 hectares) each. Thus, the small number of lots where surveyors observed these conditions is in agreement with landscape characteristics.

Witness tree analysis of the pine plains (the glacial moraine) indicates a fairly uniform composition of 58% pine (including white pine), followed by 21% white oak (Table 2-2). Pine, especially white pine, is by far the most common tree, judging by witness trees as well as surveyor remarks, which uniformly note white pine, oak, and beech as timber species on these 6 lots. Though there are fewer parcels, the composition of the witness trees in these parcels differs markedly from the composition in the landscape as a whole. In this area on Fort Drum, the presence of more disturbance adapted species such as oaks and pines, the more open understory, and the higher percentage of droughty-soil tolerant berry producing shrubs in the understory would mean greater attractiveness to Native American foraging and hunting parties.

27 Table 2-2. Witness tree composition analysis for parcels surveyed within the sandy glacial moraine.

Species Count Percent Composition White pine (Pinus strobus L.) 6 31.6 Pine (Pinus spp L.) 5 26.3 White oak (Quercus alba L.) 4 21.1 Beech (Fagus grandifolia Ehrh.) 3 15.8 Hemlock (Tsuga canadensis L., Carrière) 1 5.3 Total 19 100.0

Canfield’s Island, Lycoming County, Pennsylvania

Canfield’s Island is located in the West Branch of the Susquehanna River near the town of Williamsport, in Lycoming County, Pennsylvania (Figure 2-4). The West Branch of the

Susquehanna River flows through the Allegheny Plateau and Ridge and Valley subecoregions of

Pennsylvania, and enters the northern end of the Chesapeake Bay in Maryland. The West Branch of the Susquehanna River is in an area of the eastern United States that receives about 112 centimeters (46 inches) of rainfall per year. Average minimum and maximum temperatures range from about 4 to 14 degrees Celsius (39 to 58 degrees Fahrenheit; NOAA National Climatic Data

Center). This area is in the temperate forest biome with very high species diversity, and the dominant forest associations are dependent on topographic position. Mesophytic and hydrophilic forests, with basswood, ash, tulip-poplar (Liriodendron tulipifera L.), white oak, and maple species in the forest canopy dominate fertile valley floors, while drier oak-hickory associations can be found on uplands and side slopes, and dry oak-heath associations dominate ridgelines above the river valley. This area was glaciated during the last glacial maximum (Late

Wisconsinan glacial advance), about 22,000 years ago (Hill 2006).

28

Figure 2-4. Aerial view of the Canfield’s Island Archaeological Site, indicating the agricultural matrix the site currently lies in. The inset map shows the location of Lycoming County,

Pennsylvania.

The Susquehanna River was very important to Native Americans for transportation and the resources it provided (Kent 1984, Stewart 1993). However, despite the importance of the

Susquehanna River as a transportation waterway for people, goods, and information, European settlement of the area happened later, beginning around the late 1700s and picking up with the log boom of the mid-1800s, especially along the West Branch (Turnbaugh 1977). A major Indian

Trail went through the area, the Sheshequin Trail (Lycoming County Historical Society).

29 Resources that the rich valley provided included many important wild-gathered plants, used for both food and medicinal purposes (Appendix) and rich game resources. Because of the prevalence of high quality agricultural soils throughout the Susquehanna Valley, hunting, fishing, and gathering supplemented early companion agriculture (corn, beans, squash and tobacco were grown in the rich floodplain soils) in both the Late Woodland and Mississippian Time Periods

(Stewart 1994, Hart and Sidell 1996). The river also provided very abundant fishing, providing large quantities of protein-rich shad (Alosa sapidissima A. Wilson) and other species (Petersen et al. 1984).

Canfield’s Island is the location of the Canfield Archaeological Site (Pennsylvania

Archeological Site 36LY37). The island was created by the construction of a canal in the early

20th century for timber transportation (Lycoming County Historical Society). The whole area all along the West Branch of the Susquehanna River was cleared and homesteaded following

European settlement of the area. Agriculture was conducted during this time and into the 19th century. Agriculture was not conducted on the site after the end of the 19th century, prior to the development of the Haber-Bosch process in the early 20th century; therefore modern inorganic fertilizers have not been used on the archaeological site. The site was discovered in the year

1958, was cleared of vegetation (with minimal impact) and several excavations took place to discover the nature of the cultural resources therein. Evidence of habitation dating as far back as

4835 BC has been unearthed here, and there is a record of near-continual habitation on the site through the year 1550 AD. Cultures present on the site in the Late Woodland Period through the time of contact were the Clemson Island – Owasco, dated to 1000 AD (early Late Woodland

Period), the McFate-Quiggle complex, dated to 1550 AD (Mississippian Period), and the Shenks

Ferry culture, dated to 1456 AD (PA Historical and Museum Commission, Lycoming County

Historical Society, Stewart 1993). Tribes inhabiting the area were subgroups of the

Susquehannock Nation. The Susquehannock occupied many areas along the north-south length

30 Susquehanna River Valley, and were hostile with neighboring tribes, particularly the Lenape

(Kent 1984).

Today, the Susquehanna River flood plain is an area of rich soils, and very important for modern agricultural production. Plant growth rates in the river valley are high, as evidenced by very large specimens of species such as poison ivy (Toxicodendron radicans L., Kuntze) and grape vine (Vitis spp L.). There is very high cover of herbaceous species such as jewelweed

(Impatiens capensis Meerb.), solomon’s seal (Polygonatum biflorum Ell.), tall meadow rue

(Thalictrum pubescens Muhl.), jack-in-the-pulpit (Arisaema triphyllum L.), and bloodroot

(Sanguinaria canadensis L.), and many of these species are indicative of rich moist soils when they occur in high densities. However, because of the high level of human disturbance, and frequency of inundation carrying seed-rich sediments, the area of Canfield Island has been inundated with fast growing native and invasive exotic species such as silver maple (A. saccharinum L. - native), Japanese knotweed (Fallopia japonica Houtt.), pachysandra

(Pachysandra terminalis Siebold & Zucc.), oriental bittersweet (Celastrus orbiculatus Thunb.), honeysuckle (Lonicera spp L.) and multiflora rose (Rosa multiflora Thunb.). Many of these species have been planted as landscape ornamentals by the family living at the Canfield Estate, who previously owned much of the area on the river bank directly adjacent to Canfield Island.

Many more have been carried on the current of the river downstream from other invaded shores, and when they find a foothold on the bank, become dominant very quickly with high growth rates, high seed production, and little or no native predation.

The Canfield’s Island site was only used for soils testing and not vegetation sampling, because of modern human disturbance (present-day clearing for archaeological investigation and replanting to a prairie seed mix) and the large population of invasive plant species (the floodplain provides a continual supply of both opportunistic seeds and new seedbed for germination). In the upland areas surrounding the site, American chestnut (Castenea dentata Marsh., Borkh.) was a

31 large component of the presettlement forest (Little 1977, Mikan et al. 1994, Elliott and Swank

2008, ). The chestnut blight fungus (Cryphonectria parasitica Murrill, Barr), brought from

Japan, began to infect trees in the historical range of American chestnut around 1904. Virtually all American chestnut trees were dead (with minimal re-sprouting from unaffected root systems) by the year 1940 (Hepting 1974). Disturbance-adapted mid-successional species such as oaks and hickories replaced the lost chestnut for the most part, which precludes unbiased comparison with presettlement forest composition for witness tree analysis in this study. The area surrounding the site is also very important for agriculture, both in historical and modern times, and thus most of the land has been converted to modern agricultural fields. Therefore off-site areas were not able to be located. Baseline soils information was obtained from soil surveys of

Lycoming County, and was general in nature (i.e., no quantitative data was available on baseline soils).

Marine Corps Base Quantico, Virginia

Marine Corp Base Quantico (MCB Quantico) is located in Stafford, Fauquier and Prince

William Counties in Virginia, along the Potomac River (Figure 2-6). This is an area of rolling topography, sloping downward towards the Chesapeake Bay area and Atlantic Ocean. Major freshwater rivers of this region include the Potomac to the east and the Rappahannock to the south. Annual rainfall averages 108 centimeters (42.5 inches) per year, with 50.3 centimeters

(19.8 inches) of snowfall (NOAA National Climatic Data Center). The climate is temperate, averaging 13 degrees Celsius (55 degrees Fahrenheit). The area is unglaciated, and soils are loamy sand in texture, with loam dominating further from the Potomac River floodplain area.

Near waterways and on the eastern side of the installation, soils are sandier and well-drained.

Vegetation communities are dominated by upland hardwoods, with high importance of tulip- poplar, white oak, hickory species and ash species in areas of more moist, fertile soils. Common

32 vegetation associations in the area include oak-hickory on drier ridgelines, beech-maple on cool and moist north-facing slopes, and mixed mesophytic associations in riparian areas.

Figure 2-5. Map of Marine Corps Base Quantico indicating the location of prehistoric archaeological sites. The inset shows the location of MCB Quantico along the Potomac River in

Virginia (data on archaeological site locations supplied by the MCB Quantico CRM office).

This area on the west side of the Potomac River was inhabited by the tribes of the

Powhatan Confederacy, particularly the Patowmack or Potomac (spelling varies), and the Doegs

(sometimes called the Tauxenent, Taux or Toags), a tribe of the Algonquian Federation (Hodge

33 1912, Mook 1944). In the western part of the area could be found the Manahoac, a Siouan tribe

(Mooney 1907). Tribes of the Powhatan Confederacy as well as the Doegs had a structured societies and lived in villages, many times settling along waterways which were important for food resources such as freshwater fish and mussels, as well as transportation, communication, and trade (Stewart 1993). Much land was cleared along many navigable waterways in this region, but natives in this area may also have moved settlements inland to defensible higher ground on the tops of ridges during times of conflict (John Haynes, Cultural Resources Management at MCB

Quantico, personal communication). Although they hunted and fished, they also planted crops of maize, pumpkins (C. pepo L. variety), sunflowers, squash, beans and tobacco in fields adjacent to their villages. This area was very rich in resources, such as game and wild-gathered food and medicinal sources (Appendix), and enjoyed a longer growing season than more northern regions.

Mast species such as white and black (Q. velutina Lamb.) oaks and hickories were very important in the diet, and periodic understory burning propagated these disturbance-adapted tree species

(Day 1953). Pickerel weed (Pontederia cordata L.) and arrow arum (or tuckahoe; Peltandra virginica L., Schott) were important herbaceous sources of food, and can be found in abundance along the many slow-moving freshwater channels in this area (Figure 2-7). The Manahoac, in contrast to the Doegs, were a nomadic tribe of hunters with no established villages, who had learned to burn the forest to create grassland and attract buffalo (Bison bison Linnaeus, plains bison variety), their chief source of food (Bushnell 1926). As with other nomads, they left little evidence of their presence except for arrowheads and spear points. The Manahoac were buffered from European contact because they lived further inland from the Potomac River area than the

Doegs (Bushnell 1926, Stewart 1993).

34

Figure 2-6. Tuckahoe (utilized as important starchy food source) dominates the north side of

Chopawamsic Creek on MCB Quantico, Virginia.

In the area now occupied by MCB Quantico, Department of Defense (DoD) Cultural

Resources Management (CRM) personnel believe settlement patterns to have been typical of what is known about the lifeways of the Doeg tribe, rather than the nomadic ways of the

Manahoac. Artifacts from as far back as 9000 years before present have been located, as well as from the Woodland and transitional Mississippian Periods. The structure of the various types of habitations located has indicated two main patterns: larger villages and ceremonial centers were centralized, and single family settlements were scattered along waterways in the surrounding

35 area. Both would have had horticultural gardens associated with them. Small, single-family settlements along waterways in this area have been termed “hamlets” by Quantico CRM personnel (John Haynes, Cultural Resources Management at MCB Quantico, personal communication). Pamacocack was a larger settlement along the west side of the Potomac River between Quantico and Chopawamsic Creeks, and was identified by early European settlers’ maps of the area. This area has been developed and the site was unfortunately destroyed.

In the winter of 1607-08, the English explorer Captain John Smith traveled up the

Rappahannock River in an exploratory mission for the British crown. He sailed from Jamestown

"to the freshies" in a two-ton barge with a party of fourteen men. We have not only an account of the trip but his map of the river and its tributaries, and the location of Indian villages. His maps and records are our first written documentation of the geography and Native American tribes of the Virginia Tidewater area. Smith, regarded with suspicion by the Powhatans, sailed up from

Jamestown and found a friendly reception at "Tauxenent" on the Occoquan River, the main village of the Doeg Indians. The Doegs were members of the Algonquian Federation but, at least at that time, disliked Powhatan, which may account for the fact that they welcomed Captain

Smith with a feast at the "King's Howse" (Smith's description) which was the residence of the chief of the Doeg Tribe (Smith 1625). Smith estimated the size of the tribe to be from 135 to 170 people, including 40 bowmen. Smith also had extensive interactions and recorded descriptions of the Powhatan Confederacy and villages (Smith 1608). The arrival of the settlers in this area proved to be a disaster for the native Indians. Decimated by the Europeans' diseases for which they had no resistance and overwhelmed by the settlers' firepower in battle, they were soon driven away from the area of the modern-day Marine Corps base at Quantico (John Haynes, Cultural

Resources Management at MCB Quantico, personal communication).

The land was acquired by the United States military between the years 1917 to 1920.

Prior to that time, land usage in this area by Europeans dates back as far as the 1600s when the

36 first European settlers burned and cleared the land and built homesteads here. The Potomac River area was an important area during the Colonial Period for many reasons, including the importance of the waterway for transportation and communication, and the ready access to Atlantic coastal ports and marine resources. The land was used for agricultural purposes, and forests were cleared upon initial settlement and most likely at least once more prior to the Marine Corps acquisition of the land and the establishment of the present day second-growth forest (Beverley 1855). Cultural

Resources Managers at Quantico have identified about 220 archaeological sites, many of which are specific to the Late Woodland Time Period (i.e., contain diagnostic artifacts such as projectile points indicative of the use of bow and arrow, pot sherds indicating form and style of the Late

Woodland Period, and habitation patterns indicative of the Late Woodland). Shovel testing of many areas of the installation has been completed, which provides an opportunity to compare areas that lack cultural resources with areas where artifacts have been found.

Witness Tree Analysis

Witness tree analysis for the area of Marine Corps Base Quantico was conducted using online land deed and patent records available on the Virginia Historical Society website

(http://lvaimage.lib.va.us/LONN/LO.html; requires tiff viewer). Land Office Patents indicated the titles of the sovereign or protector under whom it was issued, the consideration for which it was issued, the name of the patentee, the size of the tract, the county of location, a description of the land, any reservations for the crown, and the date the patent was signed. No Land Office surveys are extant prior to 1779 although some county court records include survey books. This area was settled very early in American history, when compared with areas lying further from the coast (Beverley 1855). Therefore, existing land documents are those that were more carefully stored within county courthouse records for the purposes of settling property line disputes and

37 other legal issues, and these documents were many times only the land patent rather than the full survey book.

Quantico currently lies in Stafford, Prince William, and Fauquier Counties in Virginia.

However, land deed records for this analysis of witness tree data were compiled and recorded only up to the year 1700, because by this time the majority of land in this area had been cleared more than once, farmed, possibly changed hands, and the composition of the forest had generally been altered (Beverley 1855, Stewart Ware, The College of William & Mary, personal communication). The first county formed in this area of Virginia was Northumberland, formed in the year 1648. Westmoreland County was formed from it in 1653, and Stafford County was carved out of Westmoreland in 1664. As Stafford was the only county of the three that Quantico currently lies in that was formed before 1700, and Prince William and Fauquier were formed out of it in 1731 and 1759 respectively, only Northumberland, Westmoreland, and Stafford County

Land Office Patents were scoured for witness tree data. Another source of witness tree information for this area are the Northern Neck Grants. The Northern Neck is the northernmost of three peninsulas (traditionally called "necks" in Virginia) on the western shore of the

Chesapeake Bay. This peninsula is bounded by the Potomac River on the north and the

Rappahannock River on the south. Northern Neck Grants stem from a purchase of land made separately by the British crown in the year 1649 (and land grants were made from it starting in

1661), and includes land grants for Northumberland, Westmoreland, and Stafford Counties.

Northern Neck records for these counties were searched prior to the year 1700 as well. Records for Northumberland County that were searched for witness trees number 264; records for

Westmoreland County number 224, and records for Stafford County total 246.

38 Table 2-3. Witness tree counts for the area of Marine Corps Base Quantico from land patent and grant records from 1648 – 1700.

Species Count Percent Composition White oak (Quercus alba L.) 223 29.6 Red oak (Q. rubra L.) 178 23.6 Hickory (Carya spp Nutt.) 112 14.9 Oak spp (Quercus spp L.) 61 8.1 Poplar (Liriodendron tulipifera L.) 30 4.0 Pine (Pinus spp L.) 28 3.7 Gum (Nyssa spp L.) 27 3.6 Black oak (Q. velutina Lamb.) 23 3.1 Southern red oak (Q. falcata Michx.) 13 1.7 Locust (Robinia pseudoacacia L.) 11 1.5 Black walnut (Juglans nigra L.) 8 1.1 Chestnut (Castanea dentata Marsh., Borkh.) 8 1.1 Chestnut oak (Q. montana L.) 8 1.1 Beech (Fagus grandifolia Ehrh.) 6 0.8 Persimmon (Diospyros virginiana L.) 6 0.8 Maple (Acer spp L.) 4 0.5 Dogwood (Cornus spp L.) 2 0.3 Ash (Fraxinus spp L.) 1 0.1 Cedar (Juniperus virginiana L.) 1 0.1 Live oak (Quercus spp L.) 1 0.1 Mulberry (Morus rubra L.) 1 0.1 Sweetgum (Liquidambar styraciflua L.) 1 0.1 White pine (P. strobus L.) 1 0.1 Total 754 100.0

Witness trees recorded in land patents and grants prior to the year 1700 in the area of

Marine Corps Base Quantico totaled 754 trees (Table 2-3). Oak species made up the majority of the trees recorded, comprising 67 percent of the total. Hickory species were the next most recorded bearing tree at about 15 percent. This is a strong indicator that the majority of the forest composition was made up of oak species, although it is possible that a bias existed among surveyors who recorded property boundaries. However, judging by the number of species that

39 were used to mark property corners (22 total), and the fact that species that rarely reach the canopy were included (dogwood, etc), this is most likely not the case.

Later successional, disturbance-sensitive species, such as maple and beech, make up a very low percentage of the total (1.3% of total), while fire-adapted oaks and hickories and early successional species such as poplar, locust (Robinia pseudoacacia L.), and black walnut (Juglans nigra L.) were much more prominent. This indicates that the landscape was experiencing a significant amount of disturbance pre-European settlement, that can be attributed to Native

American land uses such as burning and land clearance for agriculture, especially along waterways. The vast majority of the land patents describe parcels of land that are tied to waterways, and extend back into the woods. Other information to support this that can be gleaned from these land patents and grants indicates significant presence of Native Americans in the area, with references to such features as “Indian fields”, “Indian Paths”, and “Indian Townes”.

These features are referenced very frequently; however information as to the occupants and farmers of these landscape features is lacking.

Cheatham Annex Naval Supply Station, Virginia

The Cheatham Annex Naval Supply Station (CAX) is owned by the Naval Weapons

Station at Yorktown, and lies completely within York County, Virginia. The Annex is located in the Tidewater region of coastal Virginia and is bounded by the York River on the east, Queens

Creek to the north, and Kings Creek to the south (Figure 2-8). This area has a humid temperate climate with mild winters and warm summers, and receives an average of 114 centimeters (45 inches) of precipitation per year. The average temperature ranges from 10 to 21 degrees Celsius

(50 to 70 Fahrenheit; NOAA National Climatic Data Center). The area was not glaciated during the Late Wisconsinan (Hill 2006), and has sandy, highly weathered soils (higher levels of phosphorous than glaciated soils). The topography is rolling, sloping down towards the

40 Chesapeake Bay lowlands. The Annex itself is relatively flat (slopes average less than 5%), with low bluffs along waterways, lower marshy and swampy areas along wide inlets from the York

River, and rocky beaches along larger rivers such as the York.

Figure 2-7. Map of the Cheatham Annex Naval Supply Station, York County, Virginia. The

Wilderness Area is noted with the turquoise border, and the inset notes the location of the

Cheatham Annex within the state of Virginia (data on archaeological site locations supplied by

Cheatham Annex CRM personnel).

41 Presettlement vegetation on the Annex consisted of more upland species, with a high component of longleaf pine (P. palustris Mill.) where significant burning took place (Bruce

Larson, NAVFAC Mid Atlantic Cultural Resources, personal communication; Stewart Ware, The

College of William & Mary, personal communication). However, none of the remnants of longleaf pine forests still exist today. Historical accounts indicate composition consisting of the longleaf pine as well as loblolly pine (P. taeda L.), American beech, oak species especially chestnut oak (Q. montana L.), and American holly (Ilex opaca Aiton) (Smith 1608, Smith 1625).

These associations, not including longleaf pine, are still in evidence today on the Annex (Figure

2-9). This suggests that forest composition has remained somewhat stable, despite the significant disturbances of post-European settlement in Virginia. Forest composition of sandy soils of the bluffs along Queen Creek at the north side of the Annex is dominated by loblolly pine, chinquapin (Castanea pumila Mill.), white and southern red oak (Q. falcata Michx.), beech, and holly. In the more mesic areas, forest composition is dominated by red maple, sweetgum

(Liquidambar styraciflua L.), and tulip-poplar.

42

Figure 2-8. American holly, loblolly pine, oak and beech on an archaeological site discovered through identification of shell midden evidence on Cheatham Annex, Virginia.

This area was inhabited by Algonquian-speaking peoples of the Powhatan Confederacy during the 6th century to European contact (Hodge 1912, Mook 1944). The chief Powhatan united the tribes, with the central governing body at the village of Werowocomoco on the northeast side of the York River (almost directly across the York River from the Cheatham

Annex; Gallivan 2007). These peoples farmed the land as well as hunted and gathered wild foods. They planted extensive fields of corn, beans, squash, and tobacco, starting around the year

AD 700, initiating the diagnostic lifeways of the Mississippian culture. Fields were cleared for

43 village construction and agriculture almost exclusively near waterways, while resource-gathering forays occurred throughout (Stewart 1993). They also hunted the land for small game, deer, and bear; however, the main resources utilized by these groups of people were dominated by marine resources such as shellfish harvested from brackish waterways that flow through the Tidewater zone of coastal Virginia. While hunting and gathering of wild foods and medicinals was also important, the presence of many hundreds of shellfish shells in all middens discovered indicates that resources of the brackish waterways were of the highest importance in the diet. Shell middens are very diagnostic, easily recognizable, and can be used to supplement and confirm archaeological research in this area, as well as all other coastal areas (Wyman 1868, Figure 2-10).

Figure 2-9. Shell midden evidence is present in many places along the south side of Queen Creek on Cheatham Annex, Virginia.

44

An extensive ethnobotanical inventory of Virginia has been completed as part of a project to synthesize and update existing archeobotanical data available for the entire Midatlantic region

(see McKnight and Gallivan 2007, and Appendix). This investigation of pre-maize horticulture and husbandry of estuarine plant resources has included analysis of starch grains from ceramics, stone tools, and soils. The inventory indicates that a large amount of plants that were utilized, and that Natives in this area derived use from a large variety of types of plants for food, fiber, and medicinals. The inventory includes plants that were utilized as “starvation foods” such as the inner bark of pine trees, and plants that were utilized for materials such as common mullein

(Verbascum thapsus L.), the thick woolly leaves of which were used for moccasin padding.

However, for this specific area of Virginia, there is a shorter list of plants that were utilized very frequently, because of their use in basic components of the diet, and because of their abundance on the landscape at that time. Because of the part of these plants that were collected, there is also a greater chance that seeds of these species would have germinated close to habitation areas, in middens or caches; thus they would be more useful for determining the extent of Native

American influence on the landscape. This short list includes mast of hickory species (C. cordiformis Wangenh., K Koch, glabra Miller, and tomentosa Sarg.), oak acorns (especially of the white oak family, including white oak, swamp white oak (Q. bicolor Willd.), and chestnut oak), blueberry, and paw paw (Asimina triloba L. Dunal), among several others.

Sampling on the Annex was limited to an area known as the Wilderness Area (Figure 2-8, the Wilderness Area is labeled as such) because of the lack of development. The area is used for hiking and fishing access to a small anthropogenically-created lake in the eastern section of the area, and is bounded on the north by a waterway known as Queen Creek. The Cheatham Annex

Naval Supply Station was acquired in pieces by the US Navy in the years 1918 and 1943; the

Wilderness Area was not acquired until the 1943 purchase. Cultural Resources Managers

45 employed by the Department of Defense Legacy Office have found that the soil in much of the

Wilderness Area has been impacted only by rudimentary agricultural and logging methods (i.e., mule and plow, crosscut saw and axe). Therefore the surface is relatively intact, which is a very rare occurrence in the eastern United States (Bruce Larson, NAVFAC MidAtlantic Cultural

Resources, personal communication). The land was cleared initially for homesteading and agricultural use, and some of the first European settlers lived in the area of the Annex. The interior of the Wilderness Area was impacted during the World War II era with clearing and the building of a dynamite plant by the DuPont Company, and the Patrol Road around the perimeter of the area was built in 1942. There are many large trees along the creek bank (between 200 and

300 years old according to tree cores taken in the summer of 2009), another indication that landforms in this area are relatively intact.

As archaeological testing of the Wilderness Area has not been completed, exact uses and boundaries for the three cultural sites that have been located by traditional testing methods are unknown. However, shell midden evidence all along the bank of the creek indicates extensive

Native American activity in many areas, and several sites of shell midden abundance were located. These sites may have been settlement areas with associated agricultural fields, or resource-gathering camps along the creek. It is possible that activity along the creek was limited to shell fishing, hunting, and gathering of wild resources, because of the amount of resources available in this area. Other evidence that this area contained only resource gathering camps comes from the earliest map of the area, created by Captain John Smith, which notes Indian settlements throughout the Tidewater region and coastal plain; none are located in the Wilderness

Area of the Annex (Figure 2-11). Because of the uncertainty associated with the archaeological resources in the Wilderness Area, sites located through traditional testing were compared with sites located through identification of shell midden evidence to identify characteristics they share that may be indicative of cultural significance.

46

Figure 2-10. Derivative map of the rectified version of John Smith’s map of the Chesapeake Bay area and of the English settlement at Jamestown (original version by Smith 1612; rectified version produced by Virtual Jamestown 2000); indicating the location of Werowocomoco directly across the York River from the Cheatham Annex, where no settlements were found by Smith.

Witness Tree Analysis

This area was settled very early in our nation’s history, with Capt John Smith’s expeditions to the New World for the English Crown. Much land clearance happened before

1650 for agricultural purposes. European settlement began initially in the York County area of

Virginia with the settlement of Jamestown, which catalyzed the clearance and cultivation of many

47 acres of land in the surrounding area. Land survey patents issued in Virginia were for some of the earliest lands surveyed in the New World. In 1634 lands began to be distributed among settlers with the issuance of patents. Patents indicated the titles of the sovereign or protector under whom it was issued, the consideration for which it was issued, the name of the patentee, the size of the tract, the county of location, a description of the land, any reservations for the crown, and the date the patent was signed. No Land Office surveys are extant prior to 1779 although some county court records include survey books. Virginia Land Office patents and grants have been digitized from county records, and been made available and searchable on the web.

Colonial-era forestland of the coastal Virginia area was initially bought by European settlers, and land surveyors would delineate the property corners and boundaries based on waterways, swamplands, the lands of other people, corner stakes, as well as witness trees, stumps, line trees, and sometimes even saplings. In this type of unorganized, pseudo-metes and bounds survey system, there is potentially large bias associated with which trees were used as line and boundary trees. If there is bias associated with how surveyors are choosing witness trees, a relatively small number of tree species would be included in surveys, and more rare tree species or tree species that rarely reach the canopy would generally not be included. We can reasonably assume that this type of bias is minimal in the York County surveys, as 23 species were recorded as marker trees, including typically smaller or understory species such as red mulberry (Morus rubra L.), sassafras, and persimmon (Diospyros virginiana L.). Agricultural land clearance could have occurred prior to initial land surveys, which can be indicated by the use of small trees and saplings as corners and markers; however these instances are few in number and the patterns generally follow those indicated by the data as a whole.

Land patents from court proceedings records are available beginning in the year 1623.

These are available online in an index of all the scanned patents and grants

(http://lvaimage.lib.va.us/LONN/LO.html; requires tiff viewer). For this analysis, all lands

48 surveyed from the beginning of record to the year 1700 were used for witness tree extraction. At this time, many records indicate that lands already surveyed were simply changing hands, and the length of settlement would indicate that lands had been cleared at least once, possibly multiple times, for agriculture. There are 232 total records are from York County, 112 of which include trees or stumps as boundary markers. The variety of species used (24 different species) indicates that these records include minimal surveyor bias when choosing property corner candidate trees.

The fact that most of the trees are oak species (oak, white oak, northern red oak (Q. rubra L.), black oak, water oak (Q. nigra L.), southern red oak) indicates that oak was a main component of the presettlement forest in upland areas of Tidewater Virginia. There are 285 total oaks recorded out of 502 trees total, making up 56.8% of the total composition (Table 2-4).

49 Table 2-4. Species, counts, and percentage composition in the early Colonial Period in the

Tidewater Area of Virginia, specifically York County, from land patents and grants from 1623 to

1700.

Species Count Percent Composition Red oak (Quercus rubra L.) 109 21.7 White oak (Q. alba L.) 102 20.3 Hickory (Carya spp Nutt.) 64 12.7 Pine spp (Pinus spp L.) 51 10.2 Oak spp (Quercus spp L.) 46 9.2 Poplar (Liriodendron tulipifera L.) 24 4.8 Gum (Nyssa spp L.) 24 4.8 Southern red oak (Q. falcata Michx.) 22 4.4 Maple (Acer spp L.) 11 2.2 White gum (Nyssa spp L., or Liquidambar styraciflua L.) 7 1.4 Ash spp (Fraxinus spp L.) 7 1.4 Dogwood (Cornus spp L.) 5 1.0 Chestnut (Castanea dentata Marsh., Borkh.) 5 1.0 Water oak (Q. nigra L.) 4 0.8 Persimmon (Diospyros virginiana L.) 3 0.6 Mulberry (Morus rubra L.) 3 0.6 Beech (Fagus grandifolia Ehrh.) 3 0.6 Sassafras (Sassafras albidum Nutt., Nees) 2 0.4 Cherry (Prunus spp L.) 2 0.4 Cedar (Juniperus virginiana L.) 2 0.4 Black walnut (Juglans nigra L.) 2 0.4 Black oak (Q. velutina Lamb.) 2 0.4 Myrtle (Myrtus spp L.) 1 0.2 Elm (Ulmus spp L.) 1 0.2 Total 502 100.0

Analysis of the witness tree data shows that there are four species with over 10% of the total composition each (red oak, 21.7; white oak, 20.3; hickory, 12.8; pine, 10.2). Thus, these species were the most important in the presettlement forest. All of these species are early to mid- successional, and less shade tolerant than many of the other species recorded (maple, gum (Nyssa spp L.), etc). They require a certain level of disturbance to perpetuate in a mature forest. In this

50 area, Native American people provided disturbances in the form of understory burning, clearing for agriculture and village construction, and promotion of important dietary species (such as oak and hickory) with early silvicultural techniques and mast collection.

Summary

These four sites provide an opportunity to study present day vegetation and soil charcoal patterns on identified and protected cultural sites. Many archaeological sites are on private land, or have not been fully characterized through traditional archaeological methods. Working with the DoD and CRM personnel on military bases ensures access to necessary resources, support, site records, and cultural interpretations that allow appropriate sites to be chosen for this research.

The extensive testing that has occurred at Fort Drum Army Installation, NY, and MCB Quantico,

VA, allowed suitable off-site areas to be paired with archaeological sites for comparative analyses. At Canfield’s Island Archaeological Site, PA, and the Cheatham Annex Naval Supply

Station, VA, however, suitable off-site areas were harder to locate. At Canfield’s Island, the alteration of the surrounding area precluded selection of off-site areas, while at the Cheatham

Annex, the lack of traditional archaeological testing meant that all cultural sites may not have been located. Nonetheless, all four of these sites are important locations for this research, because of the range of climate and geographic variability they cover.

51 Literature Cited

Abel, T.J. 2002. Recent research on the Saint Lawrence Iroquoians of northern New York.

Archaeology of Eastern North America 30: 137-154.

Abel, T.J. 2000. The Plus Site: An Iroquoian remote camp in upland Tompkins County, NY.

North American Archaeologist 21: 181-215.

Abrams, M.D. and Nowacki, G.J. 2008. Native Americans as active and passive promoters of

mast and fruit trees in the eastern USA. The Holocene 18: 1123-1137.

Abrams, M.D. and Ruffner, C.M. 1995. Physiographic analysis of witness-tree distribution (1765-

1798) and present forest cover through north central Pennsylvania. Canadian Journal of

Forest Research 25: 659-668.

Bamann, S., Kuhn, R., Molnar, J. and Snow, D. 1992. Iroquoian Archaeology. Annual Review of

Anthropology 21: 435-460.

Barber, M.B. and Hubbard, D.A., Jr. 1997. Overview of the human use of caves in Virginia: A

10,500 year history. Journal of Cave and Karst Studies 59: 132-136.

Beverley, R. 1855. The History of Virginia: In Four Parts. J. W. Randolph, Richmond, Virginia.

Black, B.A. and Abrams, M.D. 2001. Influences of Native Americans and surveyor biases on

metes and bounds witness-tree distribution. Ecology 82: 2574-2586.

Black, B.A., Ruffner, C.M. and Abrams, M.D. 2006. Native American influences on the forest

composition of the Allegheny Plateau, northwest Pennsylvania. Canadian Journal of

Forest Research 36: 1266-1275.

Bourdo, E.A., Jr. 1956. A review of the General Land Office Survey and of its use in quantitative

studies of former forests. Ecology 37: 754-768.

52 Brumbach, H.J. and Bender, S. 2002. Woodland Period settlement and subsistence change in the

Upper Hudson River Valley. In Northeast Subsistence-Settlement Change: A.D. 700-

1300. Hart, J.P. and Rieth, C.B., editors. New York State Museum Bulletin 496, The New

York State Education Department, Albany, New York. Pages 227-240.

Burns, R.M. and Honkala, B.H. (technical coordinators). 1990. Silvics of North America: 1.

Conifers; 2. Hardwoods. Agricultural Handbook 654, US Department of Agriculture,

Forest Service, Washington, D.C.

http://www.na.fs.fed.us/spfo/pubs/silvics_manual

Bushnell, D.I., Jr. 1926. The Indian inhabitants of the valley of Virginia. The Virginia Magazine

of History and Biography 34: 295-298.

Chapman, J., Delcourt, P.A., Cridlebaugh, P.A., Shea, A.B. and Delcourt, H.R. 1982. Man-land

interaction: 10,000 years of American Indian impact on native ecosystems in the lower

Little Tennessee River valley, eastern Tennessee. Southeastern Archaeology 1: 115-121.

Day, G.M. 1953. The Indian as an ecological factor in the northeastern forest. Ecology 34: 329-

346.

Delcourt, H.R. and Delcourt, P.A. 1996. Presettlement landscape heterogeneity: Evaluating grain

of resolution using General Land Office Survey data. Landscape Ecology 11: 363-381.

Delcourt, P.A., Delcourt, H.R., Ison, C.R., Sharp, W.E. and Gremillion, K.J. 1998. Prehistoric

human use of fire, the Eastern Agricultural Complex, and Appalachian oak-chestnut

forests: Paleoecology of Cliff Palace Pond, Kentucky. American Antiquity 63: 263-278.

Dyer, J.M. 2001. Using witness trees to assess forest change in southeastern Ohio. Canadian

Journal of Forest Research 31: 1708-1718.

Elliott, K.J. and Swank, W.T. 2008. Long-term changes in forest composition and diversity

following early logging (1919-1923) and the decline of American chestnut (Castanea

dentata). Plant Ecology 197: 155-172.

53 Foster, D.R. 1992. Land-use history (1730-1990) and vegetation dynamics in central New

England, USA. Journal of Ecology 80: 753-771.

Foster, H.T., III, Black, B. and Abrams, M.D. 2004. A witness tree analysis of the effects of

Native American Indians on the pre-European settlement forests in east-central Alabama.

Human Ecology 32: 27-47.

Fritz, G.J. 1984. Identification of cultigen amaranth and chenopod from rockshelter sites in

northwest Arkansas. American Antiquity 49: 558-572.

Gallivan, M.D. 2007. Powhatan’s Werowocomoco: Constructing place, polity, and personhood in

the Chesapeake, C.E. 1200—C.E. 1609. American Anthropologist 109: 85-100.

Gremillion, K.J. 2004. Seed processing and the origins of food production in eastern North

America. American Antiquity 69: 215-233.

Gremillion, K.J., Windingstad, J. and Sherwood, S.C. 2008. Forest opening, habitat use, and food

production on the Cumberland Plateau, Kentucky: Adaptive flexibility in marginal

settings. American Antiquity 73: 387-411.

Grimm, E.C. 1984. Fire and other factors controlling the Big Woods vegetation of Minnesota in

the mid-nineteenth century. Ecological Monographs 54: 291-311.

Hart, J.P. and Scarry, C.M. 1999. The age of common beans (Phaseolus vulgaris) in the

northeastern United States. American Antiquity 64: 653-658.

Hart, J.P. and Sidell, N.A. 1996. Prehistoric agricultural systems in the West Branch of the

Susquehanna River Basin, A.D. 800 to A.D. 1350. Northeast Anthropology 52: 1-30.

Hepting, G.H. 1974. Death of the American chestnut. Journal of Forest History 18: 60-67.

Hill, C.L. 2006. Geological framework and glaciation of the eastern area. In Handbook of North

American Indians, Volume 3: Environment, Origins, and Population. Ubelaker, D.,

Stanford, D. and Sturtevant, W.C., editors. Smithsonian Institution, Washington, D.C.

Pages 47-60.

54 Hodge, F.W. 1912. Handbook of American Indians North of Mexico, part 2 N-Z. Bureau of

American Ethnology Bulletin 30, Smithsonian Institution, Washington, D.C.

Kent, B. 1984. Susquehanna’s Indians. The Pennsylvania Historical and Museum Commission,

Anthropological Series 6, Harrisburg, Pennsylvania.

Leahy, M.J. and Pregitzer, K.S. 2003. A comparison of presettlement and present-day forests in

northeastern Lower Michigan. American Midland Naturalist 149: 71-89.

Lewandowski, S. 1987. Diohe’ko, the Three Sisters in Seneca life: Implications for a native

agriculture in the Finger Lakes region of New York State. Agriculture and Human Values

4: 76-93.

Little, Jr., E.L. 1977. Atlas of United States Trees: Minor Eastern Hardwoods, vol. 4. USDA

Forest Service Miscellaneous Publication Number 1342. United States Department of

Agriculture, Washington, D.C.

Loeb, R.E. 1998. Evidence of prehistoric corn (Zea mays) and hickory (Carya spp.) planting in

New York City: Vegetation history of Hunter Island, Bronx County, New York. Journal

of the Torrey Botanical Society 125: 74-86.

Loeb, R.E. 1987. Pre-European settlement forest composition in east New Jersey and

southeastern New York. American Midland Naturalist 118: 414-423.

Lutz, H.J. 1930. Original forest composition in northwestern Pennsylvania as indicated by early

land survey notes. Journal of Forestry 28: 1098-1103.

Marks, P.L., Gardescu, S and Seischab, F.K. 1992. Late Eighteenth Century vegetation of central

and western New York State on the basis of original land survey records. New York State

Museum Bulletin 484, The New York State Education Department, Albany, New York.

McDowell, L. 1988. Soil survey of Jefferson County, New York. USDA Soil Conservation

Service.

55 McKnight, J. and Gallivan, M. 2007. The Virginia archeobotanical database project: A

preliminary synthesis of Chesapeake ethnobotany. Archaeological Society of Virginia

Quarterly Bulletin 62: 181-189.

Mikan, C.J., Orwig, D.A. and Abrams, M.D. 1994. Age structure and successional dynamics of a

presettlement-origin chestnut oak forest in the Pennsylvania Piedmont. Bulletin of the

Torrey Botanical Club 121: 13-23.

Mook, M.A. 1944. The aboriginal population of Tidewater Virginia. American Anthropologist 46:

193-208.

Mooney, J. 1907. The Powhatan Confederacy, past and present. American Anthropologist 9: 129-

152.

National Oceanic and Atmospheric Administration (NOAA), National Climatic Data Center.

Climate Data Online website accessed July 2013.

http://www.ncdc.noaa.gov/cdo-wed

Noble, W.C. 1975. Corn, and the development of village life in southern Ontario. Ontario

Archaeology 25: 37-46.

Pearson, C.S. and Cline, M.G. 1961. Soil Survey, Lewis County, New York. USDA Soil

Conservation Service.

Petersen, J.B., Hamilton, N.D., Adovasio, J.M. and McPherron, A.L. 1984. Netting technology

and the antiquity of fish exploitation in eastern North America. Midcontinental Journal of

Archaeology 9: 199-225.

Pyne, S.J. 1983. The Indian as an ecological factor in the northeastern forest. Ecology 34: 329-

346.

Rieth, C.B. 2002. Introduction. In Northeast Subsistence-Settlement Change: A.D. 700-1300.

Hart, J.P. and Rieth, C.B., editors. New York State Museum Bulletin 496, The New York

State Education Department, Albany, New York. Pages 1-10.

56 Rieth, C.B. 2002. Early Late Prehistoric settlement and subsistence diversity in the southern tier

of New York. In Northeast Subsistence-Settlement Change: A.D. 700-1300. Hart, J.P. and

Rieth, C.B., editors. New York State Museum Bulletin 496, The New York State

Education Department, Albany, New York. Pages 209-226.

Riley, T.J., Edging, R., Rossen, J., Carter, G.F., Knapp, G., O’Brien, M.J. and Scherwin, K.H.

1990. Cultigens in prehistoric eastern North America: Changing paradigms [and

comments and replies]. Current Anthropology 31: 525-541.

Russell, E.W.B. 1981. Vegetation of northern New Jersey before European settlement. American

Midland Naturalist 105: 1-12.

Schneider, P. 1997. The Adirondacks: A History of America’s First Wilderness. Henry Holt and

Company, LLC, New York City, New York.

Schulte, L.A. and Mladenoff, D.J. 2001. The original US Public Land Survey records: Their use

and limitations in reconstructing presettlement vegetation. Journal of Forestry 99: 5-10.

Seischab, F.K. 1990. Presettlement forest of the Phelps and Gorham Purchase in western New

York. Bulletin of the Torrey Botanical Club 117: 27-38.

Seischab, F.K. and Orwig, D. 1991. Catastrophic disturbance in the presettlement forests of

western New York. Bulletin of the Torrey Botanical Club 118: 117-122.

Smith, B.D. 1989. Origins of Agriculture in Eastern North America. Science 22: 1566-1571.

Smith, B.D. 2006. Eastern North America as an independent center of plant domestication.

Proceedings of the National Academy of Sciences 103: 12223-12228.

Smith, D.G. and Crawford, G.W. 1997. Recent developments in the archaeology of the Princess

Point Complex in southern Ontario. Canadian Journal of Archaeology 21: 9-32.

Smith, J. 1624. The Generall Historie of Virginia, New England, and the Summer Isles: together

with The true travels, adventures and observations, and A sea grammer. MacLehose,

Glasgow, Scotland.

57 Smith, J. 1612. A Map of Virginia. J.Barnes, London.

Smith, J. 1608. A True Relation of Occurrences and Accidents in Virginia, 1608. In Travels and

works of Captain John Smith, President of Virginia, and Admiral of New England, 1580-

1631, Part I. John Grant, Edinburgh, published in 1910. Electronic edition in 2003 by

Ralph Bauer.

Snow, D.R. 1996. Mohawk demography and the effects of exogenous epidemics on American

Indian populations. Journal of Anthropological Archabiology 15: 160-182.

Snow, D.R. 1994. The Iroquois. Oxford: Blackwell, Cambridge, Massachusetts.

Snow, D.R. and Lanphear, K.M. 1988. European contact and Indian depopulation in the

Northeast: The timing of the first epidemics. Ethnohistory 35: 15-33.

Stewart, M. 1993. Comparison of Late Woodland cultures: Delaware, Potomac, and Susquehanna

River valleys, Middle Atlantic region. Archaeology of Eastern New York 21: 163-178.

Stewart, R.M. 1994. Prehistoric Farmers of the Susquehanna Valley – Clemson Island Culture

and the St. Anthony Site. Occasional Publications in Northeastern Anthropology, No. 13.

Moeller, R.W., general editor. Archaeological Services, Bethlehem, Connecticut.

Stothers, D.M. and Yarnell, R.A. 1977. An agricultural revolution in the Lower Great Lakes. In

Geobotany VIII. Romans, R.C., editor. Springer US, New York City, New York. Pages

209-232.

Struever, S. and Vickery, K.D. 1973. The beginnings of cultivation in the Midwest-riverine area

of the United States. American Anthropologist 75: 1197-1220.

Sullivan, R.G. and Magill, M.S. (state coordinators). 2013. NYGenWeb Project, Cayuga County.

Accessed August 2013.

http://www.rootsweb.ancestry.com/~nycayuga/land/mtracths.html

Trigger, B.G. 1978. Handbook of North American Indians – Northeast, Volume 15. Sturtevant,

W.C., editor. Smithsonian Institution, Washington, D.C.

58 Turnbaugh, W.H. 1977. Man, Land, and Time: The Cultural Prehistory and Demographic

Patterns of North-Central Pennsylvania. Published for The Lycoming County Historical

Society, Williamsport, Pensylvania, by Unigraphic, Inc., Evansville, Indiana.

Virtual Jamestown. 2000. John Smith’s Map of Virginia, 1624 Rectified: Chesapeake Bay.

Virginia Center of Digital History, University of Virginia.

http://www.virtualjamestown.org/johnsmithsmap.html

Wandsnider, L. 1997. The roasted and the boiled: Food composition and heat treatment with

special emphasis on pit-hearth cooking. Journal of Anthropological Archaeology 16: 1-

48.

Watson, P.J. 1989. Early plant cultivation in the Eastern Woodlands of North America. In

Foraging and farming: The evolution of plant exploitation. Harris, D.R. and Hillman,

G.C., editors. Unwin Hyman Ltd., London, UK. Pages 555-571.

Webster, G.S. 1984-85. Susquehannock animal economy. North American Archaeologist 6: 41-

62.

White, M.A. and Mladenoff, D.J. 1994. Old-growth forest landscape transitions from pre-

European settlement to present. Landscape Ecology 9: 191-205.

Whitney, G.G. and DeCant, J.P. 2003. Physical and historical determinants of the pre- and post-

settlement forests of northwestern Pennsylvania. Canadian Journal of Forest Research

33: 1683-1697.

Wyman, J. 1868. An account of some kjoekkenmoeddings, or shell-heaps, in Maine and

Massachusetts. The American Naturalist 1: 561-584.

Zeder, M.A., Emshwiller, E., Smith, B.D., and Bradley, D.G. 2006. Documenting domestication:

The intersection of genetics and archaeology. TRENDS in Genetics 22: 139-155.

Chapter 3

The Relationship Between Vegetation, Soil Charcoal, and

Archaeological Site Location

Introduction

Significant alteration in forest composition and structure took place throughout the time since the first people inhabited eastern North America, around 12,000 years ago (Denevan 1992,

Anderson and Moratto 1996). Estimates of Native American impacts on forests range from landscape scale (Day 1953, Pyne 1983, Delcourt 1987, Abrams and Nowacki 2008) to the local level (Russell 1983, Black and Abram 2001a). Soils may also have been impacted, both intentionally through amendments, and unintentionally through burning and midden creation

(Trigger 1978, Doolittle 1992, Mrozowski 1994). In areas such as Crawford Lake, Ontario, paleoecologists studying pollen records have noted a transition from northern hardwood- dominated to oak-pine forests types coincident with Native American habitation (1300s to 1400s;

Clark and Royall 1995). This shift to dominance of fire disturbance (indicated by a preponderance of charcoal accumulated during this time period) and fire-adapted tree species suggests that Native American burning drove changes in forest composition across a broad landscape, despite climatic influence. Delcourt et al. (1998) document significant alteration in forest composition around 3000 B.P. in Kentucky, and attribute it to Native American clearing for cultivation of native plants and broadcast understory burning. Native American land uses had broad and lasting effects in the pre-settlement landscape, and this research takes a novel approach to identifying the legacy of these impacts in the present day.

Native Americans in the eastern woodlands of North America used the forest for a wide variety of purposes and products, for food, medicinal, material, and ceremonial uses (Day 1953,

MacDougall 2003, Abrams and Nowacki 2008). Native Americans utilized specific vegetation

60 species in their diet, depending on season, climate, and proximity to fresh, brackish, or saltwater

(Fritz 1990). Around 1600 B.P. in northeastern North America, Native American groups in the eastern United States were primarily hunter-gatherers (Patterson and Sassaman 1988, Bamann et al. 1992, Custer 1994, Pihl et al. 2008), and used protoagricultural methods to cultivate semi- domesticated varieties of several native plant species (Smith 1989). The transition to the

Mississippian cultural traits was typified by the adoption of maize-based agricultural methods, however hunting and gathering continued to be very important. Land-uses included the frequent burning of the forest understory, protection and transportation of important species of plants

(Kentucky coffee tree (Gymnocladus dioicus L., K. Koch), paw paw (Asimina triloba L., Dunal);

Day 1953, Peterson 1991), girdling trees surrounding important mast-producers (Munson 1986,

Wykoff 1991, Loeb 1998), and cultivation of semi-domesticated varieties of wild plants

(MacDougall 2003, Abrams and Nowacki 2008). Other land uses, such as clearing and burning for the creation of village and camp sites and agricultural fields, indirectly promoted disturbance- adapted species, many of which were important in the diet of Native Americans (e.g., walnut

(Juglans spp L.), oak (Quercus spp L.) and other mast-producers, and blueberry (Vaccinium spp

L.) and other berry-producing shrubs; Day 1953, Abrams 1992, Black and Abrams 2001a, Foster et al. 2004).

Burning of the forest understory was a common and long-term practice of Native

American groups, as evidenced by historical records and accounts, archaeological artifacts, and paleoecological charcoal and fire scar studies (Patterson and Sassaman 1988, Dorney and Dorney

1989, Whitney 1994, Clark and Royall 1995, Delcourt and Delcourt 1997, Sutherland and

Hutchinson 2003, Black et al. 2006, Abrams and Nowacki 2008, Fesenmyer and Christensen

2010). Researchers have reported very frequent pre-European settlement fire return rates

(Guyette et al. 2002). In the northeast, where lightning ignitions (the only source of natural ignition) are relatively infrequent (Lorimer and White 2003, Kay 2007), this indicates the

61 significant influence of Native American ignitions. In eastern North America during the time period of interest to this study, understory burning facilitated travel, improved visibility in the forest, increased browse for game, increased berry and mast production, and was used to clear agricultural fields and improve hunting (Day 1953, Denevan 1992, Frost 1998). The prevalence of burning perpetuated mostly mid- and early successional tree species that are adapted to disturbances, particularly fire. This includes oak and hickory (Carya spp Nutt.) species, as well as others that are adapted to disturbance conditions (Abrams 1992, Foster et al. 1998, Dey and

Guyette 2000). Characteristics of oaks that allow them to be so well adapted to fire and other disturbances are thick bark, resprouting ability, high root to shoot ratios of growth, and the ability to compartmentalize wounds (Lorimer 1984, Abrams and Nowacki 1992, Sutherland and

Hutchinson 2003). Frequent burning leaves discernible soil charcoal that can be used as an indicator of past burning on a site (Fesenmyer and Christensen 2010).

Active fostering of woody fruit and mast species may also have been practiced by many

Native American groups. For example, the Susquehannocks and Iroquois planted black walnut

(J. nigra L.) and hickory in New York (Munson 1986, Black and Abrams 2001a). In the

Midwest, trees surrounding hickory were girdled to increase available resources and thus, mast production (Munson 1986, Wykoff 1991). This is compelling evidence of the early use of silvicultural techniques by Native Americans (Black and Abrams 2001a, Foster et al. 2004).

Other plant species may have been saved and transported when groups moved such as Kentucky coffee tree, American chestnut (Castanea dentata Marsh., Borkh), Canada plum (Prunus nigra

Aiton), groundnut (Apios americana Medikus), and leek (Allium porrum L.), although historical evidence for this is lacking and it must be confirmed through archaeobotanical testing (Day 1953,

MacDougall 2003). Native Americans in the northeastern region of the United States maintained berry patches for food and attraction of game species such as bear and deer (Day 1953, Whitney

1994, Abrams and Nowacki 2008). “Mast orchards” of oak and hickory may have been

62 maintained for the same reasons (Munson 1986, Abrams and Nowacki 2008). These patches may have been located near villages and sites of occupation for practical reasons.

Native American agriculture was important during this time in prehistory, and several native plant species were cultivated, prior to the introduction of neotropical crops such as maize

(Zea mays L.), beans (Phaseolus vulgaris L.), and subtropical squash varieties (Cucurbita spp L.;

Lewandowski 1987, Trigger 1978). The eastern United States, in fact, is believed to be the center of domestication for four wild plants, referred to as the Eastern Agricultural Complex (EAC): marshelder (Iva annua L.), chenopod (Chenopodium berlandieri Moq.), squash (C. pepo L.), and sunflower (Helianthus annuus L.), indicating both the importance of agriculture in Native

American life and the technological advancement in agriculture of which these groups of people were capable (Smith 1989, MacDougall 2003, Smith 2006). The cultivation of these species, using early methods of agriculture such as raised bed technology, included evolution through selection of varieties for various characteristics such as seed size (Smith 1989). Storage of seeds of these species was practiced throughout the harvest season, in pits dug in the ground and lined with tree bark (Wandsnider 1997, Scarry and Scarry 2005). Some of these species are still present on the landscape today, such as chenopodium or lambsquarters (C. album L.), a variety of the native cultigen, in the northeastern United States.

Vegetation species of high importance to Native Americans can be identified through ethnobotanical research. Distribution and concentration of certain plant species on the landscape, both past and present, may indicate the location of cultural resources, and where Native

Americans were active in their land uses such as burning, cultivation, and collection and storage of food such as grains and seeds. Catchment analysis of species important in the ethnobotanical record has demonstrated significant anthropological activity in the landscape of eastern North

America (Black and Abrams 2001a, Foster et al. 2004). Marks, Gardescu, and Seischab (1992) identified old Native American clearings and reported oak, hickory, and pine (Pinus spp L.)

63 growing in areas previously inhabited by peoples of the Iroquois Confederacy in central New

York. Whitney and DeCant (2003) suggested that Iroquois may be associated with oak-chestnut forests in their witness tree study across northwestern Pennsylvania, including the Allegheny

Plateau. Additional research supports the high incidence of important mast-producing and disturbance adapted species on sites previously inhabited by Native Americans (Dorney and

Dorney 1989, Black and Abrams 2001a, Black et al. 2006).

Evidence suggests that land use legacies and vegetation species propagated by Native

Americans still exist on present-day landscapes. Post-European settlement land uses perpetuated this vegetation composition in many areas (Abrams 1992). Significant land use by European settlers occurred at varying times throughout eastern North America, falling around the early

1700s in northern New York (Snow 1996), and early 1600s in coastal Virginia (Smith 1608).

European settlers would have initiated clearing of forests in much the same way as Native

Americans did, for both single and several-family settlements on the frontier (Fuller et al. 1998).

However the pace and extent of forest clearing has accelerated over time (Rhemtulla et al. 2007,

Schulte et al. 2007). These land uses would have served to continue the dominance of and even expand the disturbance-adapted species that were already prevalent in the forest due to Native

American land uses, such as oak, hickory, pine, and old-field herbaceous species, such as chenopodium.

Objectives

The overarching objective of this research is to ascertain if areas of cultural resources where significant Native American activity has occurred during the Late Woodland and

Mississippian Time Periods can be identified through the dominant vegetation composition and soil features of the present-day landscape. A general hypothesis is that Native Americans left a legacy of their land uses in the present-day ecosystem by propagating certain important wild plant

64 species and crop species in the forest, known as “indicator species”. Study of indicator species is based on the idea that the presence and number of certain plant species, including trees, shrubs, and herbs, can be an indicator of preferential utilization by Native Americans. Indicator species are discernible in the ethnobotanical record because of their prevalence in the Native diet and because the plant parts utilized (and thus gathered, cached, and transported) included the seed or seed coat of the plant. Native Americans also may have left a legacy in the soil through repeated burning that produced copious amounts of soil charcoal that can still be identified.

The specific objectives are to identify indicator species characteristic of each study area based on Native American dietary, medicinal, material, and ceremonial plant species and land uses. With these potential indicator species in mind, the current vegetation characteristic of both archaeological sites and paired off-site areas will be compared, to determine if individual indicator species or aggregated groups of indicator species (e.g., total mast species importance) can predict the presence of cultural resources on the landscape. Also, this study will determine if soil charcoal presence can be used to predict the location of archaeological sites as a legacy of

Native American burning in the soil.

Methods

In 2008-2010, an ethnobotanical literature review was undertaken for the areas of Fort

Drum, MCB Quantico, and the Cheatham Annex. Literature was reviewed to locate potential indicator species of Native American activity, and included journal articles, conversations with cultural resources management personnel, archival data, and field reconnaissance of important archaeological sites of known habitation type and length. In particular, species of importance in the diet of the Native Americans were considered to have strong potential as indicators.

Complete results of the search are contained in the Appendix of this document.

65 During the months of June-August of 2008 through 2010, each study area was visited multiple times for vegetation survey and soil charcoal analysis of cultural sites and associated non-cultural, off-site areas. Cultural sites at each military installation have been located by

Cultural Resources Management (CRM) teams, and archaeological excavations have identified the boundary of each site, the possible use of each site, and the time period the site was inhabited.

Sites with clear boundaries, minimal disturbance of on-site vegetation, and inhabited in the Late

Woodland and Mississippian time periods were chosen for survey. Off-site areas were chosen in association with each surveyed cultural site (where possible). These sites were paired with surveyed cultural sites, and located adjacent to the cultural site on the same landform, to control for soil type, geology, topography, and land-use history. Off-site areas had all tested negative for physical cultural artifacts, and it’s important to note that these methods employ both positive and negative archaeological evidence. However, just because these sites tested negative for cultural artifacts, in some cases may not necessarily mean that Native Americans did not have an effect on the landscape here. Off-site areas were difficult to find in some cases because, on military-owned lands, archaeological testing is conducted much of the time because military engineers would like to construct buildings or training sites in a certain area. If cultural resources are located by the testing, the area of the site is protected from military activities; however, any surrounding areas that test negatively are free to be used for military activities. The vast majority of the time these activities involve at least extensive clearing and ground disturbance, if not construction of buildings and impermeable surfaces.

At Fort Drum, New York, 16 archaeological sites were available for survey of vegetation and soils (50 plots total), and 8 off-site areas were located (25 plots total; Table 3-1). At MCB

Quantico, Virginia, 9 archaeological sites were surveyed (28 plots total), and 5 paired off-site areas were located (20 plots total; Table 3-2). In the Wilderness Area on the Cheatham Annex,

Virginia, only a small portion of the land had been tested for archaeological resources. It was

66 tested because it was in the “footprint” of a planned boat launch for recreational purposes. All the shovel test pits that were put in were positive for artifacts. Three separate Late Woodland sites were identified within this tested portion of the Wilderness Area, although no further archaeological testing was conducted. These sites were surveyed for vegetation composition in conjunction with 2 outside of the Wilderness Area, for a total of 5 confirmed archaeological sites

(20 plots total; Table 3-3, top section). Along the bank of the creek where the boat launch was planned, shell midden evidence was found in abundance in several areas, and is considered diagnostic of Native American habitation and influence (Waselkov 1987, Bruce Larson,

NAVFAC Mid Atlantic Cultural Resources, personal communication). The same shell midden evidence was also present further along the creek bank to the southwest of the tested area. Areas where shell midden evidence were located, but had not been tested for archaeological resources using traditional methods, were surveyed using the same methods as for all other known archaeological sites and off-site areas (7 sites, for a total of 23 plots surveyed; Table 3-3, lower section). The sites identified through shell midden evidence were considered archaeological sites even though they have not been officially tested using shovel test pits. Because of the lack of traditional testing and development that has occurred outside of the Wilderness Area, no paired off-site areas could be located. At each location, physical landscape characteristics were determined through GIS data including elevation models from which slope and aspect grids were created using ArcGIS software (ESRI 2010).

67

Table 3-1. Physical characteristics of archaeological and off-site areas at Fort Drum, New York.

Soil characteristics of FDP 1242 and 1244 were not available through Natural Resources

Conservation Service spatial data, thus textures were determined from the soil samples taken at the sites themselves (these textures are indicated with ‘*’). Sites where a suitable off-site analog for comparison were not able to be found are indicated by a ‘-‘.

Slope (degrees) Aspect Elevation (meters) Soil Series Soil texture Site Arch. Off-Site Arch. Off-Site Arch. Off-Site Arch. Off-Site Arch. Off-Site Turtle Cairn Site 1.8 0.8 northeast northwest 181.8 179.5 Galway (very stony) Galway (very stony) loam loam Boat Building Site 6.0 4.5 south west 195.4 195.5 Plainfield Plainfield sand sand Camp Drum 1 0.8 3.1 northwest northeast 161.6 168.6 Plainfield Plainfield sand sand Calendar Site 0.5 - northwest - 211.5 - Plainfield - sand - FDP1021 2.0 - north - 224.6 - Plainfield - sand - FDP1090 1.5 - northwest - 180.7 - Deerfield (loamy) - fine sand - FDP1242 10.2 - south - 229.5 - no soils info. - *silt loam - FDP1244 6.1 - west - 233.1 - no soils info. - *silt loam - FDP1161 3.0 2.6 southeast southeast 206.8 207 Plainfield Plainfield sand sand FDP1267 2.6 0.1 east northwest 213.6 213 Plainfield Plainfield sand sand FDP1268 3.5 4.8 southwest west 188.3 182 Deerfield (loamy) Deerfield (loamy) fine sand fine sand FDP1266 0.5 1.1 northwest west 180.1 180 Deerfield (loamy) Galway fine sand silt loam FDP1210 4.7 - northwest - 142.5 - Hudson and Vergennes - silt loam - FDP1272 0.7 0.8 northwest northwest 192.9 192.1 Plainfield Plainfield sand sand ASOS Site 0.7 4.5 east northeast 180.6 177 Amenia Nellis loam loam

Table 3-2. Physical characteristics of archaeological and off-site areas at MCB Quantico,

Virginia.

Slope (degrees) Aspect Elevation (meters) Soil Series Soil texture Site Arch. Off-Site Arch. Off-Site Arch. Off-Site Arch. Off-Site Arch. Off-Site 1-1 7.2 5 south south 6.7 6.5 Sassafras Sassafras fine sandy loam fine sandy loam 1-2 7.2 5 south south 8.3 6.5 Kempsville (gravelly substratum) Sassafras fine sandy loam fine sandy loam 1-3a 3 5 west south 7.9 6.5 Kempsville (gravelly substratum) Sassafras fine sandy loam fine sandy loam 1-3b 5.5 10 southeast southeast 10 25 Aura-Galestown-Sassafras complex Aura-Galestown-Sassafras complex fine sandy loam fine sandy loam 1-4 5.2 5 south south 6.7 6.5 Kempsville (gravelly substratum) Sassafras fine sandy loam fine sandy loam 1-5/1-6 6.3 8 south southwest 11.9 20 Alluvial Alluvial alluvial alluvial 2-2 4.1 3 south northwest 11 10 Wichkam Tetotum fine sandy loam fine sandy loam 4-2 4.1 3 southwest northwest 9.2 10 Wichkam Tetotum fine sandy loam fine sandy loam 6B 5.8 2 south south 38.3 33 Caroline Caroline fine sandy loam fine sandy loam

68 Table 3-3. Physical characteristics of sites discovered through traditional testing and through shell midden evidence on the Cheatham Annex, Virginia. Natural Resources Conservation Service spatial soils data for York County, VA is lacking information on the Annex. Therefore, soil textures of all samples were determined from samples taken in the field (Soil Texture column attributed with a ‘*’).

Traditional Slope (degrees) Aspect Elevation (meters) *Soil texture Boat launch 2.57 north 5.97 loamy fine sand Cellar 1.10 south 6.57 loamy fine sand Site #3 0.72 south 6.13 loamy fine sand B-8 1.30 east 2.72 fine sandy loam C-5 3.16 southeast 6.77 loamy fine sand Shell midden 1 7.65 northwest 4.61 fine sandy loam 2 5.25 west 5.36 fine sandy loam 3 4.08 northwest 7.43 fine sandy loam 4 9.44 northeast 7.48 fine sandy loam 5 8.47 northwest 15.78 fine sandy loam 6 15.90 west 16.71 fine sandy loam 7 8.40 north 16.90 fine sandy loam

Within the CRM-designated boundaries of each cultural site, stratified circular fixed-area vegetation plots were set on transects with a bearing dictated by the shape and spatial orientation of the cultural site. Generally, plot centers were at least ten meters apart, although in some sites, size and shape of the cultural site precluded such regularity. At least three plots were located within the boundary of each site. Overstory trees were identified to species and diameter at breast height (dbh) was recorded in an 800 ft2 (74.3 m2) plot. Small tree species (such as serviceberry,

Amelanchier spp Medik., and paw paw) were identified and counted in a 100 ft2 (9.3 m2) plot, and herbaceous and shrub species were quantified in a 50 ft2 (4.6 m2) plot. Herbaceous species, shrubs, and vines were given a cover class percentage within the 50 ft2 plot (0-5%, 5.1-25%,

25.1-50%, 50.1-75%, 75.1-95%, 95.1-100%), or simply recorded as present within each plot.

69 Seedling and sapling data, though taken, was not deemed suitably representative of indicator species distribution in the present-day forest because of the differing land uses acting on the forest during the early to mid-1900s, such as fire suppression, general alteration in disturbance regimes, and human activities such as harvesting. Therefore, the seedling and sapling layers in a given site were either representative of the overstory, or were the product of a changing disturbance regime influenced by post-European settlement activities. As such, they were not considered representative of cultural significance or indicator species.

Soil charcoal samples were taken in the A horizon at each vegetation plot center. The appropriate depth to Native American occupation was determined through consultation with

Cultural Resources Management personnel. At Fort Drum, this depth is ~20 cm; slower decomposition in this relatively cold climate area inhibits soil buildup. At Quantico and the

Cheatham Annex this depth is ~30 cm. When each sample was taken, the top 10 cm of soil and the organic layer were discarded. Samples were taken with a hand trowel or soil corer, placed in plastic bags, and transported back to a lab. In the lab, samples were spread on a surface to dry, and thoroughly searched for coarse soil charcoal, defined as charcoal pieces greater than 2 mm in dimension (length or width; Tryon 1948, Ohlson and Tryterud 2000, Gavin et al. 2003). Coarse soil charcoal is indicative of on-site fire, and is distinguished from fine soil charcoal by size. Fine soil charcoal is defined as charcoal pieces 1.9 mm or less in dimension and could be deposited on site through wind action from outside sources. Thus, fine soil charcoal cannot be assumed to have originated in a fire on that location. Soil charcoal samples were determined to be either positive or negative for coarse charcoal.

Data collected on overstory individuals of tree species were analyzed for frequency, density, dominance, relative frequency, relative density, relative dominance, and relative importance value. The frequency for each tree species is expressed as the number of plots in which overstory individuals of that species occurred. The density of each tree species was the

70 number of overstory individuals of that species found in all plots. Density was then scaled up to the number of trees per hectare of each species. Dominance was expressed as the total basal area for overstory individuals of each tree species across all plots, and was also scaled up to basal area

(m2) per hectare. Then the relative indices were calculated by dividing each value for each individual species by the total for all species. Finally, a relative importance value was calculated for each species by adding the relative frequency, relative density, and relative dominance for each tree species and dividing that number by three.

Analyzing data at the individual site level was somewhat confounded by low plot numbers at some sites due to archaeological site size, and the lack of directly paired off-site areas for some archaeological sites. Therefore individual site-level data were aggregated across all archaeological sites and off-site areas within each military base, using individual plots as replicates (importance values were calculated for each plot and archaeological and off-site plot averages were compared). Importance values were calculated on a species by species basis across all archaeological sites versus off-site areas, as well as for shell midden sites on the

Cheatham Annex. At Fort Drum and Quantico, overstory tree importance data was compared between archaeological sites and off-site areas. At the Cheatham Annex, data were compared between archaeological sites discovered using traditional testing methods, and archaeological sites discovered through identification of shell midden evidence in the soil. Because of the lack of off-site areas for comparison, and classification of all sites as archaeological sites, the sites discovered using the two different methods were compared to determine our ability to discern the location of archaeological sites using standard characteristics of forest composition such as prevalent species and/or presence of indicator species. Thus, if the same composition is noted between these two site types, the overstory composition could be considered typical of archaeological sites. If data for a species was distributed normally and archaeological data and off-site data displayed equal variance, a Student’s t-test was performed. If assumptions of

71 normality and equal variance were not met, a more conservative Welch’s t-test was performed.

All tests were conducted in statistical program R (R Core Team 2012).

To further elucidate patterns between archaeological and off-site plots, additional variables of interest not amenable to traditional t-tests were incorporated (such as soils and soil charcoal, etc). Classification and regression tree models employing the Random Forest (Breiman

2001, Liaw and Wiener 2002) algorithm were used to determine the importance of various species (presence/absence and quantitative measures), suites of species, and other landscape level factors in predicting the presence of archaeological significance on a given site. Random Forest is an ensemble classifier statistical method, meaning that multiple models are used to identify which of a set of categories each new observation, or the next observation, belongs. In the statistical package randomForest, if the response variable is a factor (categorical variable), classification trees are used. If the response variable is continuous, regression is used. Both involve a random subset of data that is used to build a model, or tree. To build each tree, a random sample is drawn from the original data set, and the values of a subset of the predictor variables are used to grow a classification or regression tree. The variable selection for each split in the tree (at what are known as nodes) is conducted from a small random subset of predictor variables (the number of variables to use is specified by the user; a subset of five of the predictor variables was used here). A user-specified number of trees were grown (1,000 trees were grown for each response), each with the randomly chosen subset of samples, and random selection of predictor variables. Each individual tree outputs a variable that it has found to be most able to predict the status of the response variable; each time a tree predicts a variable, it is considered a

“vote” for that variable. From the complete forest the status of the response variable is predicted as an average or majority vote of the predictions of all trees. The data used to construct the trees are known as the “in bag” sample, while the unused data are known as the “out of bag” (oob) sample. To estimate the accuracy of the predictions of the trees, the out of bag sample is run

72 through each tree, and the difference in predicted importance of each variable from the initial growth of the tree and the growth of the tree with the oob sample is known as the oob error estimation.

The datasets for classification tree analysis were built with the following variables: Site

Name (identifying name of the archaeological site or its off-site counterpart), Plot number,

Treatment (archaeological or off-site), elevation, aspect (transformed into a linear variable using the formula 1 - cos((aspect-30)*pi/180)), slope, soil series (excluding indicators of more specific characteristics such as slope), soil texture, presence/absence variables for all native species of any significance in the ethnobotanical record (i.e., species were included if they are native and generally edible or useful, regardless of their significance in literature review), importance of indicator tree species (oak, hickory, and fruit-producing trees; by number of occurrences), importance of berry-producing indicator species (both tree and shrub species, by number of occurrences), total number of utilized species surveyed per plot, and presence of coarse soil charcoal in each plot. Using this dataset, classification tree models were run using Treatment

(archaeological or off-site) as the response variable, and all other variables were used as predictors in each model. These models were run for Fort Drum and MCB Quantico vegetation data to compare overall composition and characteristics of archaeological sites versus off-site areas.

Many characteristics of the archaeological sites were not easily quantifiable or testable through traditional survey and statistical methods. These include small catchments of particular species and overall site factors that were not quantified in the field, such as site openness.

Qualitative descriptions were recorded when these types of characteristics were present on a site, and evaluated to identify patterns.

73 Results and Discussion

Identification of Indicator Species

Indicator species were identified for each area as a subset of all species utilized as food, material, or medicinal sources (Tables 3-4, 3-5, and 3-6). To qualify as an indicator species of particular importance, there must be significant reference to the species in historical accounts as being important for the diet or important medicinally. Indicator species identified at Fort Drum included oak species, which are very important mast-producers (Abrams and Nowacki 2008;

Table 3-4). Materials sources that were important are pine species, most notably pitch (P. rigida

Mill.), white (P. strobus L.), and red pine (P. resinosa Sol., Aiton) (Appendix). The proximity of this area to the Black River, an important thoroughfare, would indicate that dugout canoes would be used for transportation (Laurie Rush, Fort Drum CRM, personal communication). These canoes would have been built from pine tree trunks (Rogers 1965). Several of these tree species are also indicative of past disturbance regimes that promote fire-adapted, mid-successional species, most notably oak and pitch pine. Black cherry (Prunus serotina Ehrh.) and serviceberry were important berry-producing species in this area, and also can be indicative of site types that were ideal for habitation. Black cherry is present in areas that are slightly richer, indicating more mesic soil conditions; however in this area black cherry is not usually present in hydric or water- logged soil conditions (Burns and Honkala 1990). This type of area would be most likely near a stream or source of water, an important feature when considering camp or habitation placement.

Serviceberry, on the other hand, can be found in drier areas with more sandy soils. In this area of upstate New York, serviceberry can be found in association with oak and pine species, further adding to indicator potential in the form of a species association composed of important utilized species.

Important understory indicator species are blueberry, raspberry (Rubus spp L.), lambs quarters, mayapple (Podophyllum peltatum L.), and indian cucumber root (Medeola virginiana

74 L.). Blueberry was especially useful to Native Americans, throughout its range. There are many ethnohistoric accounts of the use of blueberries, throughout the year and in a great variety of dishes and variations (Arnason et al. 1981, Hummer 2013).

Table 3-4. Fort Drum Army Installation, NY indicator species and uses.

Species Uses White oak (Quercus alba ) acorns ground for flour, food source in winter months Northern red oak (Quercus rubra ) acorns soaked to remove tannins and ground for flour White pine (Pinus strobus ) materials source, indicate significant disturbance regime Red pine (Pinus resinosa ) indicate disturbance regime and used for boat building Pitch pine (Pinus rigida ) materials source, indicate significant fire history Black cherry (Prunus serotina ) utilized as a source of fruit in mid to late summer Blueberry (Vaccinium spp) heavily utilized as a source of fruit, patches tended Blackberry (Rubus spp) utilized as a source of fruit Wild strawberry (Fragaria virginiana ) utilized as a source of fruit Serviceberry (Amelanchier arborea ) utilized as a source of fruit Lambs quarters (Chenopodium album ) grain source, actively cultivated and transplanted Mayapple (Podophylum peltatum ) fruits eaten, root used to make a topical ointment Indian cucumber (Medeola virginiana ) root eaten raw or cooked, whole plant harvested in fall

Indicator species identified at Quantico also included important mast-producing species such as oak and hickory (Table 3-5). However the location of this area along an important waterway, the Potomac River, facilitated a different culture than that of the Iroquois further north.

Resources obtained from the Potomac were abundant fish, freshwater mussels, and several aquatic herbaceous plants whose roots were important sources of fiber for the Native Americans, most notably tuckahoe (Peltandra virginica L., Schott, Havard 1895). Habitation sites were much of the time close to the waterway and its tributaries, where rich soils favored species such as white oak (Q. alba L.) and paw paw. However, on higher bluffs surrounding these areas, sandy soils create an ideal site for mixed-oak and hickory stands, with abundant blueberry and serviceberry in the understory. In these upland areas, white pine was also an important source of

75 materials, and is indicative of significant disturbance regimes including repeated understory burning (Rogers 1965).

Table 3-5. Marine Corps Base Quantico, VA indicator species and uses.

Species Uses White oak (Quercus alba ) acorns ground for flour, food source in winter months Northern & southern red oak (Quercus spp) acorns soaked to remove tannins and ground for flour Hickory (Carya spp) several species utilized as sources of mast White pine (Pinus strobus ) materials source, indicate significant disturbance regime Black cherry (Prunus serotina ) utilized as a source of fruit in mid to late summer Paw paw (Asimina triloba ) utilized as a source of fruit, possibly transplanted Blueberry (Vaccinium spp) heavily utilized as a source of fruit, patches tended Blackberry (Rubus spp) utilized as a source of fruit Serviceberry (Amelanchier spp.) utilized as a source of fruit Spicebush (Lindera benzoin ) utilized as a medicinal and to brew tea Mayapple (Podophylum peltatum ) fruits eaten, root used to make a topical ointment Tuckahoe (Peltandra virginica ) plant harvested and all parts boiled/dried and eaten

At the Cheatham Annex, several important mast-producers were present in the presettlement forest overstory, and in high abundance (Table 3-6). These included white oak, southern red oak (Q. falcata Michx.), chinquapin (C. pumila Mill.), beech (Fagus grandifolia

Ehrh.), and hickory. However here several fruit-producing species were also prevalent, as well as freshwater resources such as tuckahoe, and higher populations of mussels in the more brackish water of the Tidewater area. The Cheatham Annex itself and the Wilderness Area in particular are situated on relatively high bluffs along waterways such as Queen Creek (see Chapter 2 of this document, figure 2-8), where sandy soils allow species such as oak, pine, blueberry, and serviceberry to proliferate.

76 Table 3-6. Cheatham Annex Naval Supply Station, VA indicator species and uses.

Species Uses White oak (Quercus alba ) acorns ground for flour, food source in winter months Southern red oak (Quercus falcata ) acorns ground for flour, food source in winter months Chinquapin (Castanea pumila ) utilized as a source of mast Beech (Fagus americana ) utilized as a source of mast Hickory (Carya spp) utilized as sources of mast Loblolly pine (Pinus taeda ) materials source, bark edible in emergencies Blueberry (Vaccinium spp) utilized as a source of fruit, patches tended to ensure supply Paw paw (Asimina triloba ) utilized as a source of fruit, possibly transplanted Persimmon (Diospyros virginiana ) utilized as a source of fruit Spicebush (Lindera benzoin ) utilized as a medicinal and to brew tea Serviceberry (Amelanchier spp) utilized as a source of fruit Blackberry (Rubus spp) utilized as a source of fruit Tuckahoe (Peltandra virginica ) plant harvested and all parts boiled/dried and eaten

Comparison of Archaeological and Off-site Overstory Composition

At Fort Drum in upstate New York, overstory tree species composition is typical of the northern hardwood forest type, including high prevalence of American beech, birch (Betula spp

L.), sugar maple (Acer saccharum Marsh.) and other maples, eastern hemlock (Tsuga canadensis

L., Carrière), and rich-site species such as tulip-poplar (Liriodendron tulipifera L.) (Table 3-7).

In this northern hardwood forest matrix, inclusions of oak, pine, and other disturbance-adapted and drier-site species can be indicative of land use history causing setbacks to the successional process that perpetuate mid-successional disturbance adapted tree species. Native American land uses, especially the use of fire in an area where lightning ignitions were infrequent, were prevalent enough to cause these changes in forest composition (Black and Abrams 2001a, Black et al. 2006).

At Fort Drum, NY, significant differences in importance values exist between several species on archaeological sites and off-site areas, including red maple (A. rubrum L.), red pine, paper birch (B. papyrifera Marsh.), scarlet oak (Q. coccinea Muenchh.), quaking aspen (Populus

77 tremuloides Michx.), and basswood (Tilia americana L.), which were more prevalent on off-site areas than archaeological sites (p < 0.05; Table 3-7). White ash (Fraxinus americana L.), black

(A. nigrum Michx.) and silver maple (A. saccharinum L.), and common buckthorn (Rhamnus cathartica L.) were all significantly more prevalent on archaeological sites than off-site areas (p <

0.05). When significant differences at p values less than 0.1 were considered, white oak and white pine are significantly more prevalent on archaeological sites than off-site areas (p values of

0.07 and 0.08, respectively). Many significant differences at p values less than 0.05 were present because the species was only discovered on one site, or one type of site, and the majority of these differences are occurring between species that are not considered cultural indicators, or important in the ethnobotanical record. An example is the higher prevalence of common buckthorn on archaeological sites. Buckthorn is not native to the eastern United States, which highlights the fact that there could be post-European land uses affecting these results. It is important to note the higher prevalence of white oak and white pine on archaeological sites, versus the higher prevalence of species such as red maple and paper birch on off-site areas. Another important result from this analysis are lower tree densities (stems/ha, Table 3-7) occurring on archaeological sites (2024) versus off-site areas (2471). This indicates that continued use of these areas by

Native Americans (and possibly subsequently by early European settlers) may have kept them more open to the present day.

78 Table 3-7. Compiled overstory data for archaeological sites (“Arch.”) and off-site areas (“Off-

Site”) at Fort Drum, NY. First columns under each heading indicate raw data, second columns indicate relative data. Significant differences between importance values on archaeological sites versus off-site areas at an alpha level of 0.05 (p < 0.05) are indicated by [**] next to the species name, while significant differences at an alpha level of 0.1 (p < 0.1) are indicated by a [*]. n = 50 archaeological site plots and 25 off-site plots

Density (stems/ha) & Rel. Dominance (m2/ha) & Rel. Frequency (# of plots) & Density Dominance Rel. Frequency Importance Value & RIV Species Arch. Off-Site Arch. Off-Site Arch. Off-Site Arch. Off-Site Sugar maple 401.8 19.9 464.8 18.8 26.6 23.8 22.8 22.4 17 17.0 11 14.5 60.6 20.2 55.7 18.6 Northern red oak 371.4 18.4 464.8 18.8 20.7 18.5 21.6 21.3 15 15.0 11 14.5 51.9 17.3 54.6 18.2 White oak* 288.0 14.2 269.1 10.9 12.1 10.8 6.8 6.7 15 15.0 7 9.2 40.0 13.3 26.8 8.9 Black cherry 197.1 9.7 146.8 5.9 10.9 9.8 8.4 8.3 7 7.0 5 6.6 26.5 8.8 20.8 6.9 White pine* 181.9 9.0 146.8 5.9 17.9 16.0 4.1 4.0 12 12.0 7 9.2 37.0 12.3 19.2 6.4 Eastern hemlock 121.3 6.0 134.5 5.4 6.1 5.5 6.4 6.3 4 4.0 3 3.9 15.5 5.2 15.7 5.2 Red maple** 106.1 5.2 354.7 14.4 4.8 4.3 8.0 7.9 6 6.0 9 11.8 15.5 5.2 34.1 11.4 Eastern hophornbeam 75.8 3.7 85.6 3.5 0.7 0.6 1.2 1.1 4 4.0 4 5.3 8.4 2.8 9.9 3.3 American beech 53.1 2.6 24.5 1.0 1.2 1.1 1.4 1.4 3 3.0 2 2.6 6.7 2.2 5.0 1.7 Mockernut hickory 45.5 2.2 48.9 2.0 1.8 1.6 3.3 3.3 3 3.0 2 2.6 6.9 2.3 7.9 2.6 White ash** 37.9 1.9 - - 1.7 1.6 - - 4 4.0 - - 7.4 2.5 - - Red pine** 30.3 1.5 122.3 5.0 1.6 1.4 8.0 7.9 2 2.0 5 6.6 4.9 1.6 19.4 6.5 Black maple** 30.3 1.5 - - 1.9 1.7 - - 2 2.0 - - 5.2 1.7 - - Silver maple** 30.3 1.5 - - 2.7 2.4 - - 1 1.0 - - 4.9 1.6 - - Serviceberry 22.7 1.1 48.9 2.0 0.2 0.2 0.3 0.3 1 1.0 3 3.9 2.3 0.8 6.2 2.1 Green ash 7.6 0.4 - - 0.1 0.1 - - 1 1.0 - - 1.5 0.5 - - Paper birch** 7.6 0.4 36.7 1.5 0.6 0.6 2.9 2.9 1 1.0 2 2.6 1.9 0.6 7.0 2.3 Black walnut 7.6 0.4 12.2 0.5 0.1 0.1 0.1 0.1 1 1.0 1 1.3 1.4 0.5 1.9 0.6 Common buckthorn 7.6 0.4 - - 0.1 0.0 - - 1 1.0 - - 1.4 0.5 - - Scarlet oak** - - 61.2 2.5 - - 4.3 4.2 - - 2 2.6 - - 9.3 3.1 Quaking aspen** - - 24.5 1.0 - - 1.7 1.7 - - 1 1.3 - - 4.0 1.3 Basswood** - - 24.5 1.0 - - 0.4 0.3 - - 1 1.3 - - 2.7 0.9 TOTAL 2024 100 2471 100 112 100 102 100 100 100 76 100 300 100 300 100

At Marine Corps Base Quantico in Virginia, the dominant forest type is mixed hardwood, with high prevalence of oak-pine forests on sandy bluffs, and the addition of American beech, sweetgum (Liquidambar styraciflua L.), and tulip-poplar in mesic areas and along waterways

(Table 3-8). White oak, chestnut oak (Q. montana L.), blackjack oak (Q. marilandica Muenchh.), and bitternut hickory (C. cordiformis Wangenh., K. Koch) had significantly higher importance

79 values on archaeological sites than off-site areas (alpha level of 0.05). Black walnut and flowering dogwood (Cornus florida L.) were significantly more prevalent on archaeological sites at an alpha level of 0.1. Slippery elm (Ulmus rubra Muhl.) was more prevalent on off-site areas

(alpha level = 0.05), and tulip-poplar was just over the significance level (p-value = 0.066) with higher importance on off-site areas. The most notable differences between archaeological sites and off-site areas are in the amount of white oak on archaeological sites (importance value of 91 on archaeological sites and 30 on off-site areas; Figure 3-8), and the amount of tulip-poplar and red maple on off-site areas (although red maple presence is more variable among sites, precluding a statistically significant comparison).

The majority of the species that were more prevalently found on archaeological sites are important sources of food for Native Americans of this area (more important when hunting and gathering activities were the main source of sustenance), and were also perpetuated through disturbances such as fire and clearing (Day 1953, Abrams 1992, Clark and Royall 1995, Foster et al. 2004). These species include oaks, hickories, and walnut. This is very notable, and indicates that these species are candidates for indicator vegetation to allow Cultural Resources Managers to more effectively locate archaeological sites in this area. This represents correlation, not causation, but is a trend with this data. Again here as at Fort Drum, archaeological sites show decreased tree densities (1682 stems/ha, Table 3-8) when compared with associated off-site areas

(1857 stems/ha), leading to the conclusion that archaeological sites may have remained more open throughout the time since European Contact.

80 Table 3-8. Compiled overstory data for archaeological sites (“Arch.”) and off-site areas (“Off-

Site”) at MCB Quantico, VA. First columns under each heading indicate raw data, second columns indicate relative data. Significant differences between importance values on archaeological sites versus off-site areas at an alpha level of 0.05 (p < 0.05) are indicated by [**] next to the species name, while significant differences at an alpha level of 0.1 (p < 0.1) are indicated by a [*]. n = 28 archaeological site plots and 20 off-site plots

Density (stems/ha) & Rel. Dominance (m2/ha) & Frequency (# of plots) & Density Rel. Dominance Rel. Frequency Importance Value & RIV Species Arch. Off-Site Arch. Off-Site Arch. Off-Site Arch. Off-Site White oak** 451.7 26.9 174.9 9.4 81.1 42.5 18.6 10.6 24 21.2 8 10.3 90.6 30.2 30.2 10.1 Sweetgum 163.4 9.7 201.8 10.9 15.4 8.1 26.2 14.8 12 10.6 8 10.3 28.4 9.5 36.0 12.0 American beech 144.2 8.6 215.3 11.6 3.4 1.8 11.4 6.5 9 8.0 6 7.7 18.3 6.1 25.8 8.6 American hornbeam 144.2 8.6 40.4 2.2 1.0 0.5 1.0 0.6 7 6.2 3 3.8 15.3 5.1 6.6 2.2 Mockernut hickory 124.9 7.4 134.5 7.2 8.3 4.3 7.5 4.3 10 8.8 4 5.1 20.6 6.9 16.6 5.5 Chestnut oak** 105.7 6.3 13.5 0.7 14.3 7.5 0.3 0.2 3 2.7 1 1.3 16.5 5.5 2.2 0.7 Northern red oak 105.7 6.3 134.5 7.2 31.4 16.5 13.6 7.7 10 8.8 8 10.3 31.6 10.5 25.2 8.4 Tulip poplar* 76.9 4.6 349.8 18.8 16.7 8.7 71.3 40.4 5 4.4 13 16.7 17.7 5.9 75.9 25.3 Red maple 76.9 4.6 349.8 18.8 1.1 0.6 7.7 4.4 8 7.1 12 15.4 12.2 4.1 38.6 12.9 Blackgum 48.1 2.9 121.1 6.5 0.8 0.4 2.7 1.6 5 4.4 6 7.7 7.7 2.6 15.8 5.3 Virginia pine 48.1 2.9 26.9 1.4 3.2 1.7 3.3 1.9 2 1.8 2 2.6 6.3 2.1 5.9 2.0 Black oak 38.4 2.3 26.9 1.4 6.4 3.4 6.7 3.8 4 3.5 2 2.6 9.2 3.1 7.8 2.6 Flowering dogwood* 38.4 2.3 - - 0.4 0.2 - - 3 2.7 - - 5.1 1.7 - - American holly 28.8 1.7 13.5 0.7 0.5 0.2 0.1 0.1 3 2.7 1 1.3 4.6 1.5 2.1 0.7 Pignut hickory 28.8 1.7 13.5 0.7 1.9 1.0 0.1 0.1 3 2.7 1 1.3 5.4 1.8 2.1 0.7 Blackjack oak** 19.2 1.1 - - 1.8 0.9 - - 1 0.9 - - 3.0 1.0 - - Black walnut* 9.6 0.6 - - 0.2 0.1 - - 1 0.9 - - 1.6 0.5 - - Southern red oak 9.6 0.6 13.5 0.7 0.9 0.5 5.0 2.9 1 0.9 1 1.3 1.9 0.6 4.9 1.6 Bitternut hickory** 9.6 0.6 - - 1.9 1.0 - - 1 0.9 - - 2.5 0.8 - - Ash species* 9.6 0.6 - - 0.3 0.2 - - 1 0.9 - - 1.6 0.5 - - Slippery elm** - - 13.5 0.7 - - 0.6 0.3 - - 1 1.3 - - 2.3 0.8 Eastern redcedar - - 13.5 0.7 - - 0.1 0.0 - - 1 1.3 - - 2.0 0.7 TOTAL 1682 100 1857 100 191 100 176 100 113 100 78 100 300 100 300 100

Forest overstory composition in York County, Virginia, the location of the Cheatham

Annex Naval Supply Station, is similar to that of MCB Quantico. The mixed hardwood forest type dominates, with shifts in prevalence of certain species depending on soil moisture conditions. Tulip-poplar and sweetgum, and cherry to a certain extent, dominated in more mesic

81 conditions, while oak, hickory, and loblolly pine (P. taeda L.) dominated in the majority of the

Wilderness Area along sandy bluffs above Queen Creek. The majority of archaeological sites were located in the Wilderness Area, and contained a higher importance of xeric-adapted species that were important in the Native American diet.

At the Cheatham Annex, VA, tulip-poplar, northern red oak (Q. rubra L.), and loblolly pine were well represented in cultural sites discovered both through traditional methods and shell midden evidence (Table 3-9). Significant differences exist in importance of sweetgum, eastern red cedar (Juniperus virginiana L.), and black walnut (alpha level of 0.05; Table 3-9). White oak and hickory species were exclusively found on sites discovered with shell midden evidence (oaks significantly different at p-values of 0.00 and 0.01, hickory at a p-value of 0.08). This suggests that the southern shore of Queen Creek was most likely a suitable site for Native American resource gathering activity. The primary driving force in Native American habitation was likely the presence of food resources in the Creek itself. In addition, food and resource gathering activities occurred further inland in the areas where the oak and hickory species thrived on the drier, sandier soils of the river bluffs. The rather ubiquitous use by Native Americans of both riparian and upland sites in this area makes it difficult to elucidate differences in vegetation composition between cultural and non-cultural sites. The various ways in which this landscape was used may have perpetuated cultural indicator species throughout the area, and indeed many important mast and fruit producing species are located close along the shore of Queen Creek, including white oak, Allegheny chinquapin, and persimmon (Diospyros virginiana L.). This area does not show shell midden evidence, however it was most likely heavily traversed. Overstory tree densities are much higher on sites discovered with shell midden evidence (1521 stems/ha) than on sites discovered through traditional testing (1101 stems/ha, Table 3-6). I believe this is because traditional testing took place in an area planned for a boat launch, which was more low- lying and periodically flooded, and thus had more understory vegetation and less overstory trees

82 than shell midden sites (refer to Table 3-3). The differences seen between these two types of archaeological sites provide further evidence for ubiquitous use of the area, regardless of vegetation characteristics.

Table 3-9. Compiled overstory data for archaeological sites (“Arch.”) and sites identified through shell midden evidence (“Midden”) at Cheatham Annex Naval Supply Station, VA. First columns under each heading indicate raw data, second columns indicate relative data. Significant differences between importance values on archaeological sites discovered through traditional testing versus those discovered with shell midden evidence at an alpha level of 0.05 (p < 0.05) are indicated by [**] next to the species name, while significant differences at an alpha level of 0.1 (p

< 0.1) are indicated by a [*]. n = 20 traditional archaeological site plots and 23 shell midden site plots

Density (stems/ha) & Rel. Dominance (m2/ha) & Frequency (# of plots) & Density Rel. Dominance Rel. Frequency Importance Value & RIV Species Arch. Midden Arch. Midden Arch. Midden Arch. Midden Sweetgum** 269.1 24.4 - - 13.0 15.9 - - 7 20.6 - - 60.9 20.3 - - Eastern redcedar** 171.2 15.6 46.8 3.1 7.6 9.2 0.6 0.5 6 17.6 2 4.1 42.4 14.1 7.6 2.5 Tulip poplar 146.8 13.3 140.4 9.2 29.5 36.0 23.8 17.1 5 14.7 4 8.2 64.0 21.3 34.5 11.5 Northern red oak 146.8 13.3 140.4 9.2 15.1 18.5 13.1 9.4 3 8.8 5 10.2 40.6 13.5 28.8 9.6 Flowering dogwood 73.4 6.7 117.0 7.7 1.0 1.2 0.9 0.6 3 8.8 3 6.1 16.7 5.6 14.5 4.8 Black walnut** 73.4 6.7 - - 10.6 12.9 - - 3 8.8 - - 28.4 9.5 - - Loblolly pine 73.4 6.7 70.2 4.6 2.0 2.5 10.8 7.8 2 5.9 2 4.1 15.0 5.0 16.5 5.5 Black cherry* 48.9 4.4 - - 0.4 0.5 - - 1 2.9 - - 7.9 2.6 - - Common hackberry** 24.5 2.2 - - 1.8 2.2 - - 1 2.9 - - 7.3 2.4 - - Black locust 24.5 2.2 - - 0.1 0.2 - - 1 2.9 - - 5.3 1.8 - - American beech** 24.5 2.2 117.0 7.7 0.6 0.7 13.8 9.9 1 2.9 3 6.1 5.9 2.0 23.7 7.9 Sassafras* 24.5 2.2 - - 0.2 0.2 - - 1 2.9 - - 5.4 1.8 - - Swamp white oak** - - 163.8 10.8 - - 16.7 12.0 - - 6 12.2 - - 35.0 11.7 Pignut hickory* - - 70.2 4.6 - - 0.6 0.5 - - 2 4.1 - - 9.2 3.1 American holly* - - 234.0 15.4 - - 2.2 1.6 - - 6 12.2 - - 29.2 9.7 Red maple** - - 163.8 10.8 - - 15.5 11.1 - - 6 12.2 - - 34.1 11.4 Mockernut hickory - - 23.4 1.5 - - 0.1 0.1 - - 1 2.0 - - 3.7 1.2 White oak** - - 234.0 15.4 - - 40.8 29.4 - - 9 18.4 - - 63.1 21.0 TOTAL 1101 100 1521 100 81.97 100 139 100 34 100 49 100 300 100 300 100

83 Comparison of Archaeological and Off-site Soil Charcoal

Across all study sites, the presence of soil charcoal on and off cultural sites was somewhat variable (Tables 3-10, 3-11, and 3-12). At Fort Drum, charcoal was present in a greater number of samples from cultural sites than from off-site areas (Table 3-10), however when soil samples from an archaeological site were positive for soil charcoal, samples from the off-site area tended to be positive as well. Most archaeological sites surveyed at Fort Drum had at least 50% of samples positive for soil charcoal (11 out of 15 cultural; Table 3-10). However, six of nine off-site areas on Fort Drum had at least 33% of samples positive for coarse soil charcoal, indicating burning in these locations as well. At MCB Quantico, six of nine archaeological sites had 50% or greater positive samples compared with five of nine with at least

20% for the associated off-site area (Table 3-11). The presence of coarse soil charcoal in many off-site areas, albeit reduced relative to cultural sites, may suggest that Native American burning

“spilled over” into the adjacent off-site areas used in this study. This analysis has potential to aid in delineation of a “sphere of influence” of any Native American village or significant habitation site. Alternatively, this could represent the effect of later post-European settlement burning of a wider area, as vertical movement of soil charcoal in the soil profile could skew these results.

84 Table 3-10. Soil charcoal results for archaeological and off-site areas on Fort Drum, NY.

Samples were considered positive for soil charcoal if they contained coarse soil charcoal (>1 mm in diameter), and sites were considered positive if at least 50% of samples contained coarse soil charcoal. A ' - ' indicates that no suitable off-site area was located for comparison with the cultural site, due to military activities and/or construction.

Site Arch. Control Turtle Cairn Site 50 33 Boat Building Site 66 100 Camp Drum 1 33 0 Calendar Site 71 - FDP1021 0 - FDP1090 100 - FDP1242 100 - FDP1244 100 - FDP1161 60 75 FDP1267 100 100 FDP1268 50 0 FDP1266 0 0 FDP1210 0 - FDP1272 100 100 ASOS Site 50 33

85 Table 3-11. Soil charcoal results for archaeological and off-site areas on MCB Quantico, VA.

Samples were considered positive for soil charcoal if they contained coarse soil charcoal (>1 mm in diameter), and sites were considered positive if at least 50% of samples contained coarse soil charcoal. A ' - ' indicates that no suitable off-site area was located for comparison with the cultural site, due to military activities and/or construction.

Site Arch. Control 1-1 100 20 1-2 33 20 1-3a 100 20 1-3b 100 33 1-4 66 20 1-5/1-6 50 0 2-2 50 0 4-2 0 0 6B 0 0

Soil charcoal results at the Cheatham Annex, Virginia, were considered differently than those from Fort Drum and Quantico (Table 3-12). At Cheatham Annex 10 of the 12 study sites were located within the Wilderness Area (see Chapter 2 of this document, figure 2-8), and all excavations tested positive for Late Woodland/Mississippian cultural activity. Therefore, all locations were either archaeological sites or midden sites. Over half of the sites sampled (7 of 12 cultural sites; Table 3-12) were positive for soil charcoal evidence, and all of these occurred within the Wilderness Area. These sites were dominated by oaks, hickories, and loblolly pine, all fire-tolerant and disturbance-adapted species. This indicates that although archaeological evidence is present throughout, the presence of soil charcoal is highly correlated with the presence of many important indicator species. Vegetation and soil charcoal data in conjunction

86 may increase the ability of Cultural Resources Management personnel to discern differing land use patterns in areas of ubiquitous Native American activity.

Table 3-12. Soil charcoal results for archaeological sites discovered through traditional testing and through identification of shell midden evidence on Cheatham Annex, VA. Samples were considered positive for soil charcoal if they contained coarse soil charcoal (>1 mm in diameter), and sites were considered positive if at least 50% of samples contained coarse soil charcoal. A ' -

' indicates that no suitable off-site area was located for comparison with the cultural site, due to military activities and/or construction.

Site Arch. Control Traditional Boat Launch 0 - Cellar 33 - Site #3 0 - B-8 0 - C-5 0 - Shell Midden 1 67 - 2 100 - 3 100 - 4 0 - 5 100 - 6 100 - 7 50 -

Classification Tree Modeling

Classification tree models were run for the two study sites where archaeological versus off-site areas could be identified, Fort Drum, New York, and MCB Quantico, Virginia (Figures 3-

1 and 3-2). The response variable for the classification models was treatment type

(archaeological or off-site). Classification tree model results are reported here as importance

87 values of predictor variables. Importance values are expressed as the percent increase in mean square error (MSE) of the model when the predictor variable in question is eliminated from the model (mean decrease in accuracy of the model). An importance value of 10% would therefore indicate that MSE of a model built without the predictor variable in question would have an MSE value 10% higher than MSE of a model built including that predictor variable. Variables can be considered informative if their importance value is greater than the absolute value of the lowest negative-scoring predictor variable, because the importance values of irrelevant variables vary randomly around zero (Strobl et al. 2009, Shih 2011). Classification tree model results from Fort

Drum reveal that trees per hectare in each plot was the most significant of the predictors at an importance value of 24.1% (Figure 3-1). This supports the observation in the analysis of overstory composition of archaeological sites versus off-site areas that archaeological sites tend to be more open. Elevation, aspect, soil series, and slope follow closely behind with importance values ranging from 21.2% to 19.0%. Presence/absence of all vegetation species range in importance from 10.4% (serviceberry) to -2.3% (beech; negative importance values represent a decrease in model error when the variable is removed as a predictor, indicating an increase in accuracy of the model and no relationship with the response variable). Important tree and berry- producing species both had importance values of about 9%, total number of indicator species recorded was at 7.8%, and soil charcoal had an importance of 3.4%. Many predictor variables were relevant to the model, including Indian cucumber root presence, blueberry presence, and soil charcoal presence. The absence of maple was also revealed to be a relevant predictor for treatment type, with a higher incidence of overstory maple on off-site areas than archaeological sites.

88

Figure 3-1. Importance values, or mean decrease in accuracy of the model, for predictor variables in a classification tree model predicting Treatment (archaeological or off-site area) for data collected at Fort Drum.

Trees per hectare and landscape variables are indicated as having the highest importance in predicting treatment type (archaeological or off-site). This may indicate that using geotopographical characteristics in conjunction with vegetation would be the best predictor of the potential for cultural resources. Fort Drum Cultural Resources Management personnel have

89 employed a GIS model with some success in predicting site locations (Ft Drum GIS personnel,

Terrace Model; see Zeidler 2001, Crawford and Smith 2002). This model used a topographic index that predicted the location of what can be characterized as terrace or bench sites, areas of minimal slope occurring above the elevation of stream valleys. With the relative success of this model, factors that influence or are influenced by topography should be important. However the model used by Drum CRM was used to broadly identify areas of potential cultural resources of any time period, while this study is concerned with a particular group of Native Americans during a relatively constrained time period.

At Quantico, landscape-level geomorphological factors were again the most important factors predicting the response variable of treatment type. Importance values for plot-level trees per hectare, slope, aspect, elevation, and soil series ranged from 23.1% (soil series) to 16.1%

(trees per hectare; Figure 3-2). Decreased tree densities on archaeological sites (Table 3-8) underpin the importance of stems per hectare in locating cultural landscapes. Importance values of the presence/absence of vegetation species ranged from 13.3 (white oak, a significant indicator species) to -2.4 (raspberry). Important indicator species occurrences and total indicator occurrences had relatively low importance values (0.3 to 5.6%), however total indicator species present and importance mast-producing tree species present did have importance values above

2.4%, indicating a small measure of relevance to prediction of treatment type (archaeological or off-site area). Soil charcoal was a somewhat better predictor of treatment, with an importance value of 9.0%. As at Fort Drum, geological and topographical variables were the most important, leading to similar hypotheses regarding the relationship between these predictor variables and archaeological resources across the broader landscape of MCB Quantico. However at MCB

Quantico, white oak appears to show importance values fairly close to landscape position variables, which is a very significant finding for this research.

90

Figure 3-2. Importance values, or mean decrease in accuracy of the model, for predictor variables in a classification tree model predicting Treatment (archaeological or off-site area) for data collected at MCB Quantico.

The presence of white oak had the highest importance among the vegetation presence/absence variables as a predictor of treatment type. This is consistent with observations by John Haynes, former Cultural Resources Manager for MCB Quantico, that archaeological sites often seem to be located in areas of high white oak importance in the overstory (personal communication). Indeed, cultural sites in certain areas of the base have high white oak

91 importance (Table 3-8) which indicates that this trend is correlated with archaeological significance. White oak is a calciphile and thus is most often found in areas of specific soil chemistry, often characteristic of topographical positions such as valley, bench and cove sites

(Abrams 2003). Soil testing of sites containing higher incidence of white oak in the overstory indicated increased calcium and magnesium (parts per million), and slightly higher pH (see

Chapter 4 of this document, Table 4-3), and soil series was also a good predictor of treatment type (with an importance value of 23.1%). These lines of evidence suggest a correlative relationship between soil type, white oak presence, and archaeological resources.

Qualitative consideration of the presence or absence of certain indicator species was a very useful analysis, when trends in vegetation on archaeological sites or off-site areas were not fully or accurately characterized by traditional quantitative methods. An example is the distribution of chenopodium (or lambsquarters) at Fort Drum, a native grain and early cultigen in the Eastern Agricultural Complex (Smith 1984, Smith 2006). At Fort Drum, chenopodium is found in open areas on archaeological sites and off-site areas on the sandy glacial outwash moraine, as well as areas to the east of the moraine. These areas were not captured in many vegetation plots because they often occur in more open sites and on the periphery of archaeological sites. However, this species was noted in reconnaissance surveys of six important archaeological sites (the Calendar Site, Camp Drum 1, and Fort Drum Prehistoric sites 1161,

1267, 1093, and 1020), in vegetation surveys of two important sites (FDP 1267, FDP 1021), and in the surrounding areas of the open, oak savanna type vegetation on the sandy glacial moraine.

Another trend noted at Fort Drum was the characteristic of sparse understory vegetation on archaeological sites, especially those where artifacts such as post molds and hearths were discovered (these features indicate structures and thus semi-long-term habitation). This was most notable at Camp Drum 1, the site of a St. Lawrence Iroquoian village during the Late Woodland

Time Period (Fort Drum Cultural Resources Management). Native American village sites would

92 have been good defensible positions, within close proximity to fresh water, and on dry, well- drained, and level ground (Trigger 1978). These villages were inhabited for some period of time before resources in the surrounding area were likely depleted; then the village would be moved to another location (Trigger 1978). These areas were inhabited by Native Americans in a rotation, and then by Europeans for their defensibility, proximity to freshwater, and other desirable settlement characteristics (e.g., cleared, fertile areas). The results of this study and others suggest that continuation of use, such as for agriculture or timbering, of these areas maintained them in a pre-European vegetation state (Black and Abrams 2001b, Foster and Motzkin 2003). Also noted were lower overstory tree densities on archaeological sites both at Fort Drum and Quantico, further supporting this conclusion.

At Fort Drum, archaeological sites are most prevalent in areas of sandy, well-drained soil, especially the sand plains of the glacial moraine. These characteristics make it attractive for habitation and resource-gathering, including abundant sources of food, well-drained soils, and relatively clear understory. Therefore, a specific species composition of several key species (oak, pine, blueberry, and serviceberry) is prevalent on many of the archaeological sites on Fort Drum.

This composition is indicated as being important in overstory comparison, and is correlated with soils and topography and well as archaeological resources. Similarly, at MCB Quantico many

Late Woodland cultural sites are found with a higher incidence of white oak, and on soils of a particular type supporting vegetation associated with rich, mesic areas. Although the trend of white oak presence on archaeological sites is not supported by all sites located on Quantico, many of the important sites where actual relatively long-term habitation was expected to have occurred do uphold the pattern. Other archaeological sites on MCB Quantico tended to be found in areas of more xeric soils and composition, indicated by increased importance of species such as chestnut oak. These trends could provide information to cultural resources personnel on site

93 distribution across a broad landscape, utilizing both geotopographical characteristics and indicator vegetation to elucidate settlement patterns versus seasonal use.

A potential factor influencing these results is the effect of broad landscape level practices such as understory burning, and the creation of a “sphere of influence” surrounding areas of high anthropological significance. One of the main factors of Native American land use, broadcast understory burning, has been shown to be a very important tool associated with human habitation

(Pyne 1983, Patterson and Sassaman 1988, Whitney 1994). If, as many researchers suggest, this was the driving factor in the compositional change of forests in the pre-European settlement landscape (Abrams and Nowacki 2008), it may be that the changes affected were broad and thus reduced site-level differences between archaeological sites and adjacent off-site areas. Forest clearing in wide swaths surrounding habitation sites may have contributed as well to these broader impacts, possibly making many “off-site” areas unsuitable for these comparisons.

Conclusion

At Fort Drum, New York, and MCB Quantico, Virginia, higher incidence of certain indicator species and soil charcoal are associated with cultural sites. This trend could be typified by one species (such as white oak at MCB Quantico), or it can be a grouping of several species

(such as the species association that dominates the sandy glacial moraine at Fort Drum). In areas of fairly ubiquitous usage of the landscape or a particular feature of the landscape, such as the riparian areas and bluffs above brackish waterways of the Cheatham Annex, Virginia, variation in the distribution of fire-dependent species and soil charcoal can aid in identifying what particular type of Native American usage the land may have experienced (e.g., habitation, gathering and processing of plant food resources, gathering of shellfish, etc). Decreased tree densities on archaeological sites at Fort Drum and MCB Quantico also indicate that in these landscapes, areas of sparser overstory should be prioritized for archaeological testing.

94 Pinpointing a combination of geological, soil, topographical, and vegetation characteristics that are typical of Native American cultural sites from the last several centuries could allow Cultural Resources Management programs on military lands to more efficiently identify landscapes where the probability of locating archaeological features is greater. Analysis of these features may also allow CRM personnel to more accurately characterize the types of land use or activities that occurred on these sites. Soil charcoal analysis could serve as a rapid method of determining fire history during the time period of interest, but may not aid in locating exact cultural site boundaries because of its ubiquitous nature on both archaeological and off-site areas.

Surveys of vegetation could be integrated into testing strategies either before or after archaeological site discovery, to both locate and prioritize landscapes of particular interest, or further elucidate what kinds of activities may have been occurring following the discovery of a site. The presence of white oak in the area of MCB Quantico can be viewed as an indicator of past Native American land use and habitation, while pyrogenic communities such as oak-pine- blueberry at Fort Drum, are key communities that indicate high probability of the presence of significant cultural resources. Where these species exist in combination with other utilized species and decreased tree densities, the likelihood of discovery of cultural resources increases greatly.

95 Literature Cited

Abrams, M.D. 2003. Where has all the white oak gone? BioScience 53: 927-939.

Abrams, M.D. 1992. Fire and the development of oak forests. BioScience 42: 346-353.

Abrams, M.D. and Nowacki, G.J. 2008. Native Americans as active and passive promoters of

mast and fruit trees in the eastern USA. The Holocene 18: 1123-1137.

Abrams, M.D. and Nowacki, G.J. 1992. Historical variation in fire, oak recruitment, and post-

logging accelerated succession in central Pennsylvania. Bulletin of the Torrey Botanical

Club 119: 19-28

Anderson, M.K. and Moratto, M.J. 1996. Native American land-use practices and ecological

impacts. Sierra Nevada Ecosystem Project: Final Report to Congress, volume 2, chapter

9. University of California, Davis. Pages 187-206.

Arnason, T., Hebda, R.J., and Johns, T. 1981. Use of plants for food and medicine by native

peoples of eastern Canada. Canadian Journal of Botany 59: 2189-2325.

Bamann, S., Kuhn, R., Molnar, J. and Snow, D. 1992. Iroquoian Archaeology. Annual Review of

Anthropology 21: 435-460. a. Black, B.A. and Abrams, M.D. 2001. Influences of Native Americans and surveyor biases on

metes and bounds witness-tree distribution. Ecology 82: 2574-2586. b. Black, B.A. and Abrams, M.D. 2001. Analysis of temporal variation and species-site

relationships of witness tree data in southeastern Pennsylvania. Canadian Journal of

Forest Research 31: 419-429.

Black, B.A., Ruffner, C.M. and Abrams, M.D. 2006. Native American influences on the forest

composition of the Allegheny Plateau, northwest Pennsylvania. Canadian Journal of

Forest Research 36: 1266-1275.

Breiman, L. 2001. Random Forests. Machine Learning 45: 5-32.

96 Burns, R.M. and Honkala, B.H. (technical coordinators). 1990. Silvics of North America: 1.

Conifers; 2. Hardwoods. Agricultural Handbook 654, US Department of Agriculture,

Forest Service, Washington, D.C.

http://www.na.fs.fed.us/spfo/pubs/silvics_manual

Clark, J.S. and Royall, P.D. 1995. Transformation of a northern hardwood forest by aboriginal

(Iroquois) fire: Charcoal evidence from Crawford Lake, Ontario, Canada. The Holocene

5: 1-9.

Crawford, G.W. and Smith, D.G. 2002. Early Late Woodland in southern Ontario: An update

(1996-2000). In Northeast Subsistence-Settlement Change: A.D. 700-1300. Hart, J.P. and

Rieth, C.B., editors. New York State Museum Bulletin 496, The New York State

Education Department, Albany, New York. Pages 73-96.

Custer, J.F. 1994. Current archaeological research in the Middle Atlantic region of the eastern

United States. Journal of Archaeological Research 2: 329-360.

Day, G.M. 1953. The Indian as an ecological factor in the northeastern forest. Ecology 34: 329-

346.

Delcourt, H.R. 1987. The impact of prehistoric agriculture and land occupation on natural

vegetation. TREE 2: 39-44.

Delcourt, H.R. and Delcourt, P.A. 1997. Pre-Columbian Native American use of fire on southern

Appalachian landscapes. Conservation Biology 11: 1010-1014.

Delcourt, P.A., Delcourt, H.R., Ison, C.R., Sharp, W.E., and Gremillion, K.J. 1998. Prehistoric

human use of fire, the Eastern Agricultural Complex, and Appalachian oak-chestnut

forests: Paleoecology of Cliff Palace Pond, Kentucky. American Antiquity 63: 263-278.

Denevan, W.M. 1992. The Pristine Myth: The landscape of the Americas in 1492. Annals of the

Association of American Geographers 82: 369-385.

97 Dey, D.C. and Guyette, R.P. 2000. Anthropogenic fire history and red oak forests in south-central

Ontario. The Forestry Chronicle 76: 339-347.

Doolittle, W.E. 1992. Agriculture in North America on the eve of Contact: A reassessment.

Annals of the Association of American Geographers 82: 386-401.

Dorney, C.H. and Dorney, J.R. 1989. An unusual oak savanna in northeastern Wisconsin: The

effect of Indian-caused fire. American Midland Naturalist 122: 103-113.

ESRI 2010. ArcGIS Desktop: Release 10. Redlands, California: Environmental Systems Research

Institute.

Fesenmyer, K.A. and Christensen, N.L., Jr. 2010. Reconstructing Holocene fire history in a

southern Appalachian forest using soil charcoal. Ecology 91: 662-670.

Foster, D.R. and Motzkin, G. 2003. Interpreting and conserving the openland habitats of coastal

New England: Insights from landscape history. Forest Ecology and Management 185:

127-150.

Foster, D.R., Motzkin, G., and Slater, B. 1998. Land-use history as long-term broad-scale

disturbances: Regional forest dynamics in central New England. Ecosystems 1: 96-119.

Foster, H.T., III, Black, B., and Abrams, M.D. 2004. A witness tree analysis of the effects of

Native American Indians on the pre-European settlement forests in east-central Alabama.

Human Ecology 32: 27-47.

Fritz, G.J. 1990. Multiple pathways to farming in precontact eastern North America. Journal of

World Prehistory 4: 387-435.

Frost, C.C. 1998. Presettlement fire frequency regimes of the United States: A first

approximation. In Fire in Ecosystem Management: Shifting the paradigm from

suppression to prescription. Pruden, T.L. and Brennan, L.A., editors. Tall Timbers Fire

Ecology Conference Proceeding, number 20. Tall Timbers Research Station, Tallahassee,

Florida. Pages 70-81.

98 Fuller, J.L., Foster, D.R., McLachlan, J.S., and Drake, N. 1998. Impact of human activity on

regional forest composition and dynamics in central New England. Ecosystems 1: 76-95.

Gavin, D.G., Brubaker, L.B., and Lertzman, K.P. 2003. Holocene fire history of a coastal

temperate rain forest based on soil charcoal radiocarbon dates. Ecology 84: 186-201.

Guyette, R.P., Muzika, R.M., and Dey, D.C. 2002. Dynamics of an anthropogenic fire regime.

Ecosystems 5: 472-486.

Havard, V. 1895. Food plants of the North American Indians. Bulletin of the Torrey Botanical

Club 22: 98-123.

Hummer, K.E. 2013. Manna in winter: Indigenous Americans, huckleberries, and blueberries.

HortScience 48: 413-417.

Kay, C.E. 2007. Are lightning fires unnatural? A comparison of aboriginal and lightning ignition

rates in the United States. In Proceedings of the 23rd Tall Timbers Fire Ecology

Conference: Fire in Grassland and Shrubland Ecosystems. Masters, R.E. and Galley,

K.E.M., editors. Tall Timbers Research Station, Tallahassee, Florida. Pages 16–28.

Lewandowski, S. 1987. Diohe’ko, the Three Sisters in Seneca life: Implications for a native

agriculture in the Finger Lakes region of New York State. Agriculture and Human Values

4: 76-93.

Liaw, A. and Wiener, M. 2002. Classification and regression by randomForest. R News 2: 18-22.

Loeb, R.E. 1998. Evidence of prehistoric corn (Zea mays) and hickory (Carya spp.) planting in

New York City: Vegetation history of Hunter Island, Bronx County, New York. Journal

of the Torrey Botanical Society 125: 74-86.

Lorimer, C.G. 1984. Development of the red maple understory in northeastern oak forests. Forest

Science 30: 3-22.

99 Lorimer, C.G. and White, A.S. 2003. Scale and frequency of natural disturbances in the

northeastern US: Implications for early successional forest habitats and regional age

distributions. Forest Ecology and Management 185: 41-64.

MacDougall, A. 2003. Did Native Americans influence the northward migration of plants during

the Holocene? Journal of Biogeography 30: 633-647.

Marks, P.L., Gardescu, S and Seischab, F.K. 1992. Late Eighteenth Century vegetation of central

and western New York State on the basis of original land survey records. New York State

Museum Bulletin 484, The New York State Education Department, Albany, New York.

Mrozowski, S.A. 1994. The discovery of a Native American cornfield on Cape Cod. Archaeology

of Eastern North America 22: 47-62.

Munson, P.J. 1986. Hickory silviculture: A subsistence revolution in the prehistory of eastern

North America. In Emergent Horticultural Economies of Eastern Woodlands. Keegan,

W.F., editor. Southern Illinois University, Carbondale, Illinois. Pages 1-20.

Ohlson, M. and Tryterud, E. 2000. Interpretation of the charcoal record in forest soils: Forest fires

and their production and deposition of macroscopic charcoal. The Holocene 10: 519-525.

Patterson, W.A. and Sassaman, K.E. 1988. Indian fires in the prehistory of New England. In

Holocene Human Ecology in Northeastern North America. Nicholas, G.P., editor.

Plenum Press, New York, New York. Pages 107-135.

Peterson, R.N. 1991. Pawpaw (Asimina). In Genetic Resources of Temperate Fruit and Nut

Crops. Moore, J.N. and Ballington, J.R., editors. Acta Horticulturae 290: 567-600.

Pihl, R.H., Monckton, S.G., Robertson, D.A., and Williamson, R.F. 2008. Settlement and

subsistence change in the turn of the first millennium: The view from the Holemdale Site,

Brantford, Ontario. In Current Northeast Paleoethnobotany II. Hart, J.P., editor. New

York State Museum Bulletin 512, The New York State Education Department, Albany,

New York. Pages 151-172.

100 Pyne, S.J. 1983. Indian fires. Natural History 2: 6-11.

R Core Team. 2012. R: A Language and Environment for Statistical Computing. R Foundation

for Statistical Computing, Vienna, Austria.

http://www.R-project.org

Rhemtulla, J.M., Mladenoff, D.J., and Clayton, M.K. 2007. Regional land-cover conversion in the

U.S. upper Midwest: Magnitude of change and limited recovery (1850-1935-1993).

Landscape Ecology 22: 57-75.

Rogers, E.S. 1965. The dugout canoe in Ontario. American Antiquity 30: 454-459.

Russell, E.W.B. 1983. Indian-set fires in the forests of the northeastern United States. Ecology 64:

78-88.

Scarry, C.M. and Scarry, J.F. 2005. Native American ‘garden agriculture’ in southeastern North

America. World Archaeology 37: 259-274.

Schulte, L.A., Mladenoff, D.J., Crow, T.R., Merrick, L.C., and Cleland, D.T. 2007.

Homogenization of northern U.S. Great Lakes forests due to land use. Landscape

Ecology 22: 1089-1103.

Shih, S. 2011. Random forests for classification trees and categorical dependent variables: An

informal quick start R guide. Stanford University, Stanford, Connecticutt.

www.stanford.edu/~stephsus/R-randomforest-guide.pdf

Smith, B.D. 2006. Eastern North America as an independent center of plant domestication.

Proceedings of the National Academy of Sciences 103: 12223-12228.

Smith, B.D. 1989. Origins of agriculture in eastern North America. Science 246: 1566-1571.

Smith, B.D. 1984. Chenopodium as a prehistoric domesticate in eastern North America: Evidence

from Russell Cave, Alabama. Science 226: 165-167.

101 Smith, J. 1608. A True Relation of Occurrences and Accidents in Virginia, 1608. In Travels and

works of Captain John Smith, President of Virginia, and Admiral of New England, 1580-

1631, Part I. John Grant, Edinburgh, published in 1910. Electronic edition in 2003 by

Ralph Bauer.

Snow, D.R. 1996. Mohawk demography and the effects of exogenous epidemics on American

Indian populations. Journal of Anthropological Archabiology 15: 160-182.

Strobl, C., Malley, J., and Tutz, G. 2009. An introduction to recursive partitioning: Rational,

application, and characteristics of classification and regression trees, bagging, and

random forests. Psychological Methods 14: 323-348.

Sutherland, E.K. and Hutchinson, T.F. 2003. Characteristics of mixed-oak forest ecosystems in

southern Ohio prior to the reintroduction of fire. USDA Forest Service General

Technical Report NE-299.

Trigger, B.G. 1978. Handbook of North American Indians – Northeast, Volume 15. Sturtevant,

W.C., editor. Smithsonian Institution, Washington, D.C.

Tryon, E.H. 1948. Effect of charcoal on certain physical, chemical, and biological properties of

forest soils. Ecological Monographs 18: 81-115.

Wandsnider, L. 1997. The roasted and the boiled: Food composition and heat treatment with

special emphasis on pit-hearth cooking. Journal of Anthropological Archaeology 16: 1-

48.

Waselkov, G.A. 1987. Shellfish gathering and shell midden archaeology. Advances in

Archaeological Method and Theory 10: 93-210.

Whitney, G.G. 1994. From Coastal Wilderness to Fruited Plain. Cambridge University Press,

Cambridge, UK.

102 Whitney, G.G. and DeCant, J.P. 2003. Physical and historical determinants of the pre- and post-

settlement forests of northwestern Pennsylvania. Canadian Journal of Forest Research

33: 1683-1697.

Wykoff, M.W. 1991. Black walnut on Iroquoian landscapes. Northeastern Indian Quarterly 4: 4–

17.

Zeidler, J.A. 2001. Dynamic modeling of landscape evolution and archaeological site

distributions: A three-dimensional approach. SERDP Project CS-1130. Center for

Environmental Management of Military Lands TPS 01-8, Colorado State University, Fort

Collins, Colorado.

Chapter 4

Native American Black Earth in the Eastern United States

Introduction

Aboriginal man had the ability to amend and improve the fertility of inherently infertile soils to stimulate agricultural production, increasing levels of soil organic matter and nutrients

(Pape 1970, McFadgen 1980, Dudal 2002, Blume and Leinweber 2004, Hesse and Baade 2009).

Human-modified soils have been reported in many areas all over the world (Mücher et al. 1990,

Jiménez-Osornio and Gomez-Pompo 1991, Bokhorst et al. 2005, Kögel-Knabner et al. 2010). In the Brazilian Amazon Basin, 2000 year old Terra Preta (or ‘black earth’) soils received human amendments over hundreds of years (Mann 2000, Lehmann et al. 2003). Important amendments to soils included smothered burning of agricultural and human waste, as well as plant biomass

(low or no oxygen prevents combustion). This creates a substance that has been termed “biochar”

(Hecht 2004, Steiner et al. 2007). These human-related activities increase the amount of calcium, carbon, phosphorous, and nitrogen and darkened the soil strata (Skinner 1986). This enabled the nutrient-deplete tropical soils to support agriculture for large advanced civilizations (Mann 2000,

Lehmann et al. 2003). Native groups of highland Mesoamerica also formed raised fields to which additional nutrient-rich muck and other organic material were added. This practice was a component of the chinampa system (Jácome 1993). The Native Americans inhabiting eastern

North America may also have amended soils, especially in areas of inherently low soil fertility

(Mrozowski 1994).

Maize-based (Zea mays L.) agriculture was introduced in the northeastern U.S. around

AD 700-1000; thisvaries depending on location (Lewandowski 1987, Struever and Vickery 1973,

Loeb 1998). However, hunting and gathering remained important, and the subsistence methods that typified the Late Woodland culture (a combination of maize-based agriculture and hunting

104 and gathering) extended up until the time of first contact in many more northern areas of the eastern U.S. (Fritz 1990). Prior to the introduction of maize, several plants were brought under human cultivation; maygrass (Phalaris caroliniana Walter), lambs quarters (Chenopodium berlandieri Moq.), knotweed (Polygonum erectum L.), sunflower (Helianthus annuus L.), and marsh elder (Iva annua L.), however these plants were more typical of the mid- to southeastern

U.S. (sites in Virginia and North Carolina), while the most notable in the northeast was lambs quarters (New York; Day 1953). Agricultural systems were not advanced, however, until the introduction of maize and techniques associated with “three sisters” cultivation. This introduction and the following societal shifts were so profound as to mark the beginning of the

Late Woodland Time Period.

Agricultural systems during this time were thought to be typified by slash and burn techniques (Sykes 1980, Bamann et al. 1992, Doolittle 2004). The intended agricultural site would be burned, after which the ash and charcoal produced by the burn would have been left on the field as an amendment (Trigger 1978, Delcourt 1987). The creation of agricultural fields was typically labor intensive, often requiring girdling of the trees on a site and clearing of all vegetation (Doolittle 1992). For this reason, it would have been in the best interest of the Native

American groups to increase the useful life of the fields through soil amendments with charcoal, human waste, and organic refuse. These amendments would have darkened and increased the fertility of soils, conditions that may persist to the present day. Alternatively, soils would have received little attention beyond application of the charcoal from initial burning, and would have been depleted of nutrients in 7-15 years (e.g., Butzer 1990, Duane Quates, Fort Drum archaeologist, personal communication). The depleted field would have been allowed to lie fallow in a swidden agricultural system, and another area would be prepared. This type of system would be described as extensive rather than the intensive Terra Preta agricultural system.

Extensive agricultural systems may have been more prevalent in areas of higher baseline soil

105 fertility (i.e., floodplains; Doolittle 2004), or where group territories had to be large enough to support the number of deer required for making clothing. Another way Black Earth could have been indirectly created was through the formation of middens, or refuse piles that would inevitably form through build-up of waste from a settlement in one spot for an extended period of time (“Indian dirt,” Woods 1984, Laurie Rush, Fort Drum CRM, personal communication).

There is historical evidence that some Native groups in the eastern U.S. were using agricultural amendments at the time of Contact with European explorers and settlers (Mrozowski

1994). Much of this evidence comes from ethnohistoric accounts (Jesuit missionaries, French fur trappers, early English and Spanish explorers, etc.; Rountree 1989). For example, the people of the Waumpanoag group in the coastal northeast are said to have applied fish remains to agricultural fields as an amendment (Morozowski 1994). However, some anthropologists are skeptical, questioning the practicality of using a main food source, such as fish, as fertilizer (Ceci

1975, Ceci 1990, Hurt 2002). Alternate hypotheses exist as to the nature of Native agriculture in eastern North America: some experts believe that agriculture was primarily performed in an extensive nature due to the abundance of land area (Butzer 1994, Hurt 2002, John Haynes, MCB

Quantico CRM, personal communication). However, in areas of higher Native American population density, such as coastal plains and along waterways, is it logical to believe that competition would be occurring among groups for the best agricultural soils. In this case, fields would probably be farmed more intensively.

Objectives

A general hypothesis is that in areas of low soil fertility (e.g., sandy glacial outwash or rocky coastal plains), amendments would have been necessary for the productive growing of horticultural plants. However, in areas of higher soil fertility amendments would not have been necessary for crop production. The research described in this chapter deals with occurrence and

106 properties of Native American black earth soils at four locations in the northeast and mid-Atlantic regions of the eastern U.S. The objectives of this research were generally to determine the present-day soil properties on archaeological sites, including nutrient levels and color, and to compare them with soil properties in areas that have tested negative for physical cultural artifacts.

The type of cultural site, whether a low intensity occupied site such as a resource gathering camp or a more intensively occupied site such as a village, was considered a crucial characteristic in the potential for discovery of human-modified soil properties.

Specific objectives were to determine the color and nutrient properties of soils on and off

Native American archaeological sites. Differences in these soil fertility patterns possibly due to the influence of baseline productivity levels of regional soils were determined. A separate methodology was undertaken to determine if length and intensiveness of occupation have an effect on soil characteristics, focusing on occupation sites with long term habitation that would have had associated horticultural or agricultural fields. Also on these intensively inhabited sites, it was determined if differences exist between sites located along waterways versus inland, possibly due to the influence of the resources the waterways could provide precluding the need for intense agricultural production.

Methods

During the summers of 2009, 2010, and 2011, the study sites were visited and soil sampling was conducted on known archaeological sites of various types and associated off-site areas. Archaeological sites at each study area were chosen for each analysis that was conducted based on characteristics of the archaeological site itself and the artifacts identified. On-site versus off-site sampling was conducted on known Native American village sites as well as well as outlying sites where the type of artifacts found does not indicate long term habitation. These sites will hereafter be referred to as “resource camps” due to the type of activity that most likely

107 occurred there: hunting, food-gathering, materials gathering (such as chert quarries), or areas where building of items such as dugout canoes occurred (Fort Drum Cultural Resources

Management). More intensive sampling techniques were used only when long term, somewhat intensive habitation could be inferred from artifacts such as post molds, indicating that semi- permanent structures were erected. Middens and hearths were not used as positive indicators of what could be termed a village or hamlet site, as these types of features would also be present on sites inhabited for a relatively short amount of time.

Sampling of known archaeological sites and nearby off-site areas was conducted at Fort

Drum and MCB Quantico during the summer months of 2008 through 2010. Archaeological site locations and descriptions were obtained from Cultural Resources Management (CRM) personnel, and sites were chosen for the study based on time period of habitation, minimal confounding factors such as habitation in later time periods, and level of current disturbance.

From three to six plots were set per archaeological site, depending on the size of the site. An equal number of paired off-site plots were set off the archaeological sites, controlling for geophysiological characteristics and topography. Soil samples were taken at plot centers set along transects (bearing dictated by size and shape of archaeological site) within each cultural site and associated off-site area. Samples were taken with a hand trowel to the depth of occupation

(~20 cm at Fort Drum and ~30 cm at Quantico and Cheatham Annex, top 10 cm and organic layer were removed). Soil color was determined in the field using the Munsell 10YR system of color coding (Figure 4-1). Samples were then sifted for organic materials, dried as soon as possible, and sealed in plastic bags. They were analyzed at the Penn State Agriculture Analytical Services

Lab for pH, phosphorous (P; ppm), potassium (K; ppm), magnesium (Mg; ppm), calcium (Ca; ppm), acidity (meq/100g), cation exchange capacity (CEC; meq/100g), percent saturation of the

CEC by K, Mg, and Ca, and analysis of trace element composition (zinc, copper, and sulfur; in ppm). Total Nitrogen (N; ppm) was also analyzed on each on and off-site sample from Fort

108 Drum and MCB Quantico. It is important to note that values of pH were tested to ascertain variability among sampled areas because many nutrients are more or less available at differing levels of soil acidity or basicity.

Soil data from the resource camp sites at Fort Drum and Quantico were initially analyzed by aggregating data obtained for all archaeological sites into one dataset, comparing it with data obtained for all off-site areas, and testing differences in means for measured soil parameters. If data for a specific soil parameter was distributed normally and archaeological data and off-site data displayed equal variance, a Student’s t-test was performed. If assumptions of normality and equal variance were not met, a more conservative Welch’s t-test was performed. All tests were conducted in statistical program R (R Core Team 2012). T-tests were used to compare mean values of pH, phosphorous, potassium, magnesium, calcium, cation exchange capacity, nitrogen, zinc, copper, sulfur and color between samples taken on cultural sites and samples taken on adjacent off-site where no cultural significance was discovered. Nutrient levels were compared in parts per million, cation exchange capacity was compared in millequivalents per 100 grams of soil (meq/100g) and color was compared using a system of continuous coding of Munsell 10YR values (Table 4-1; Munsell Color Company 1975). This coding system relies on a scaling from darkest soil to lightest soil, with allowances for the fact that colors of black and brown are more representative of soil organic matter content than colors of gray or yellow. Differences in mean values of the above nutrients and soil characteristics were also compared using the same methods between off-site samples collected at Fort Drum, and off-site samples collected at Quantico to determine if differences in inherent soil fertility exist between these two geographically removed locations.

109

Figure 4-1. Munsell 10YR (red-yellow) system of assigning color to soil samples. On the left are color descriptions, and on the right are color plates for comparison to soil samples in the field.

Table 4-1. Munsell 10YR soil color code and corresponding analysis code for statistical analysis.

Munsell Value Analysis Code 2/1 1 2/2 2 3/1 3 3/2 4 3/3 5 3/4 and 3/6 6 4/4 and 4/6 7 4/3 8 4/2 9 4/1 10 5/3 11 5/4 to 5/8 12 5/2 13 5/1 14

110 A lumped analysis of all cultural sites and off-site areas does not account for differences in soil types among cultural sites, or allow evaluation of the most important variables driving observed differences among sites. Therefore, further analysis was conducted on archaeological site versus off-site data at Fort Drum and MCB Quantico using regression tree modeling.

Regression tree modeling was used to examine variation in patterns among disparate study areas

(Fort Drum and Quantico), and among archaeological sites within each study area, and to assess correlative relationships between all variables. The model chosen for this analysis was Random

Forest (Breiman 2001), and it was run in statistical program R, using the package randomForest

(Liaw and Wiener 2002). In the statistical package randomForest, if the response variable is a factor (categorical variable), classification trees are used. If the response variable is continuous, regression is used. All response variables designated in these models are continuous, therefore regression trees were built. Variable selection at node splits within individual trees was conducted from a random subset of six of the total predictor variables, and 1,000 trees were grown for each model.

Random Forest models were run using values of pH, phosphorus, potassium, magnesium, calcium, cation exchange capacity, zinc, copper, sulfur, nitrogen, and soil color category as response variables. In each model, all variables besides that assigned as the response were used as predictor variables, as well as Military Base (Fort Drum or MCB Quantico), Site of Origin

(name of archaeological site sampled; associated off-site areas were tagged with the same Site of

Origin as archaeological sites), Site Type (archaeological or off-site), and Plot number. Models were also run using data from each individual base only; in these models, Military Base as a predictor was eliminated however all other variables remained the same. Each model predicted the most robust classifier variable of the response from the set of twelve (or eleven) predictors.

To evaluate each model, variable importance was used. Variable importance is a measure of the decrease in prediction capacity of the model when the variable considered is replaced with a

111 random variable, expressed as the percent increase in mean square error (MSE). Percent increase in MSE values was used as a measure of the correlative relationship between response and predictors, and particularly as a measure of the ability of Site Type to predict each response.

In summer 2011, at archaeological sites identified as significant long-term Native

American habitation areas, soils samples were taken in a grid to identify inclusions of higher fertility soils. Archaeological sites chosen for this portion of the analysis were limited to sites that show clear evidence of human habitation for 10+ years during the Late Woodland Time

Period, with the assumption that these sites would have had agricultural fields associated with them, mostly in the form of raised bed agriculture. Sites that do not show clear archaeological evidence of long term habitation cannot be inferred to have experienced the type of land use necessary to produce the signal of amendment in the soil. These areas include resource camps, foraging and hunting camps, tool-making or boat-making areas, and ceremonial sites. At Fort

Drum in upstate New York, a grid was placed on and surrounding a St. Lawrence Iroquois palisaded village known as Camp Drum 1 (area of about 8 hectares, Figure 4-2). In Lycoming

County, PA, a grid was placed over the central and surrounding area of the Canfield Island

Susquehannock village and ceremonial center, covering about 4 hectares (Figure 4-3). At MCB

Quantico, a grid was placed on and surrounding several hamlet sites identified along the north bank of Chopawamsic Creek; the entirety of the north bank that had not been developed was surveyed in this way, an area amounting to about 16 hectares in size (Figure 4-4). At the

Cheatham Annex along the southwest side of the York River in Virginia, a grid of soil samples was placed adjacent to the south bank of Queen Creek and extending south, including the area of two identified Late Woodland archaeological sites, and surrounding untested areas, covering about 43 hectares (shell middens were also in evidence in some of the untested area, a major indicator of Native American use and influence; Figure 4-5).

112

Figure 4-2. Location of soil samples on Camp Drum 1 Archaeological Site on Fort Drum Army

Installation, NY.

113

Figure 4-3. Location of soil samples on Canfield Island Archaeological Site in the

Susquehannock River near Williamsport in Lycoming County, PA.

114

Figure 4-4. Location of soil samples in an area of “hamlet” type archaeological sites along the north bank of Chopawamsic Creek on MCB Quantico, VA.

115

Figure 4-5. Location of soil samples in the Wilderness Area on Cheatham Annex Naval Supply

Station, VA.

The grid sampling was placed to cover the entirety of each of these archaeological sites or areas, as well as the surrounding areas. GPS waypoints were taken at each sample, and samples were extracted from the ground using a soil corer with a 25 millimeter diameter tube for minimal disturbance to cultural resources present on the sites. Off-site samples were obtained for calculation of baseline soil characteristics at Fort Drum and Quantico because of the availability of archaeological testing information for the areas surrounding the sites. This information was

116 not available for the Cheatham Annex, and at Canfield Island the surrounding area had been converted to modern agriculture, precluding the availability of unbiased off-site samples. All samples were assigned a soil color using the Munsell system in the field, and prepared using the same method described for previous samples. Samples were analyzed for the same soil characteristics and nutrient content as previous samples (aside from nitrogen levels due to monetary limitations) at the Penn State Agriculture Analytical Services Lab.

Grid samples from Fort Drum, Canfield Island, and MCB Quantico were analyzed using a spatial interpolation method in ArcGIS from ESRI (ESRI 2010). Grid samples from the

Cheatham Annex were displayed using alteration of symbology (graduated symbol size) in the program ArcGIS, because the distance between sampled areas and relatively low number of samples caused irregular patterns in spatial interpolations that were artifacts of sample placement, making this method less ideal for visualizing patterns in the data. Waypoints taken at each sample site were uploaded and converted to shapefiles using DNRGarmin (Minnesota

Department of Natural Resources 2008), and a “site boundary” was identified using either heads- up digitizing in ArcMap or through obtaining archaeological site boundary shapefiles from participating installations. Soil sample data for pH, cation exchange capacity, calcium, magnesium, potassium, phosphorus, and soil color were used to assign the value of each measured soil characteristic to the corresponding sample in the attribute table of the waypoint shapefile. Where baseline soil information was available (Fort Drum and Quantico), the mean value of each measured soil characteristic for all baseline samples was subtracted from the corresponding value of the grid sample. Positive resulting values indicate areas where the value of the grid sample exceeded the value of the baseline, indicating increased soil pH, or fertility

(nutrient levels and cation exchange capacity). ArcToolbox Spatial Interpolation Tools in

ArcMap (ArcGIS by ESRI software package; ESRI 2010) were used to develop a surface to interpolate each soil characteristic of interest across the entirety of the area encompassed by the

117 grid of soil samples at each site. Spatial Interpolation is the process of assigning values to unknown points by using values from a set of known points. Inverse Distance Weighting (IDW) was the spatial interpolation technique used. With IDW, unknown points are assigned a value equal to the weighted average of nearby sampled points, where the weight given each point is an inverse proportion to the distance to the unknown point. The interpolation is run using the field in the attribute table of the shapefile containing the information about the location of each individual grid sample. The model outputs a raster grid file with an interpolated value assigned to each individual pixel in the grid. The model parameters for the analysis conducted at Fort Drum were an output cell size of 2 meters, a variable search radius using the 10 nearest points to interpolate the value of the unknown point, and a power parameter of 2. At Canfield Island, model parameters were an output cell size of 1 meter, a variable search radius using the 10 nearest points, and a power parameter of 2. At MCB Quantico, parameters were an output cell size of 3 meters, a variable search radius using the 10 nearest points, and again a power parameter of 2. The power parameter is the exponential value that weight will decrease as distance increases; in this case moving one unit of distance will cause the weight to decrease by an exponent of 2. This parameter controls how much surrounding points influence the interpolated value. The interpolated surfaces were used to locate and interpret patterns and trends in the data, and relate them to potentially causative Native American activities. Specifically, catchment analysis will be conducted with these results to identify particular areas where data reveal a concentration of soil samples with values that may be indicative of an area potentially amended for agricultural use.

118 Results and Discussion

Comparison of Overall Soil Data from Archaeological and Off-site Areas

Archaeological sites (including resource camps and sites inhabited more intensively) at

Fort Drum displayed higher (though not statistically significant) values of essential nutrients phosphorus, potassium, magnesium, calcium and nitrogen and higher cation exchange capacity

(CEC) than associated off-site areas (Table 4-2). Soil color was significantly darker (lower analysis code value; alpha level of 0.05) than soil color in off-site areas, indicating higher prevalence of black and dark brown colors and soil fertility (increased organic matter; Franzmeier

1988). A Munsell color of dark yellowish brown was characteristic of the off-site areas, which is indicative of lower organic matter and thus, increased leaching of key nutrients and decreased soil fertility. Cultural sites also displayed higher zinc and copper, but lower sulfur. Values of pH were almost exactly the same across cultural and off-site areas. Similar pH values allowed unbiased comparisons among values of soil nutrients between the two populations, as nutrient availability can be impacted by pH (IPNI 2010).

High variability among samples precluded statistical strength on most comparison tests between cultural and off-site samples (alpha level 0.05). However, the general trend of increased nutrient status on cultural sites is notable, as are p-values below 0.1 for phosphorus, potassium and magnesium (Table 4-2). Potassium and magnesium are essential nutrients, as well as base cations, and phosphorus is a key element in plant nutrition (Mengel and Kirkby 2001). Therefore, a trend of increased values of these particular nutrients could possibly indicate an important biological, if not statistical, difference in soils of cultural versus off-site areas.

119 Table 4-2. Mean values (+/- standard error) of soil parameters of interest and p-values associated with t-tests for significant difference in means between for cultural sites and associated off-site areas at Fort Drum, NY. Soil sample depth is 20 centimeters into the A horizon. n = 32 archaeological samples and 30 off-site samples

Cultural Sites Off-Site Areas P-value pH 5.18 ± 0.12 5.18 ± 0.15 0.99 Phosphorus 49.66 ± 7.86 33.67 ± 5.22 0.094 Potassium 60.94 ± 12.82 37.53 ± 3.95 0.089 Magnesium 58.53 ± 9.60 39.47 ± 6.02 0.097 Calcium 478.28 ± 90.32 411.80 ± 92.49 0.61 CEC 9.99 ± 0.70 9.21 ± 0.72 0.44 Zinc 2.87 ± 0.31 2.05 ± 0.31 0.065 Copper 0.78 ± 0.07 0.62 ± 0.05 0.07 Sulfur 25.31 ± 3.35 25.12 ± 3.22 0.97 Nitrogen 1885.99 ± 193.99 1791.28 ± 188.97 0.73 Soil color 5.19 ± 0.56 6.70 ± 0.49 0.045

The cultural sites at Quantico displayed similar nutrient concentrations, pH, and soil color when compared with off-site areas (Table 4-3). Mean values of all soil parameters tested were not statistically different between cultural and off-site areas at an alpha level of 0.05 or 0.1 in all cases. This test indicated that soil characteristics are not indicative of the presence of archaeological resources; however variation among different sites in soil series may confound results by causing high variation among samples. This potentially indicates that patterns of soil fertility at MCB Quantico may be the result of different usage of the landscape in this area where

Natives depended largely on resources from the waterways of the Potomac River and tributaries

(such as Chopawamsic Creek; Stewart 1993, John Haynes, Quantico CRM, personal communication). Alternatively, the lack of differences between soil characteristics of archaeological sites and off-site areas may be due to inherently high soil fertility in this area, facilitating Native agriculture without the aid of human soil amendments. Another possibility is

120 that these soil samples could’ve been contaminated with other post-European settlement land use impacts in the soil, masking the effect of Native American amendments and impacts.

Table 4-3. Mean values (+/- standard error) of soil parameters of interest and p-values associated with t-tests for significant difference in means between for cultural sites and associated off-site areas at MCB Quantico, VA. Soil sample depth is 30 centimeters into the A horizon. n = 28 archaeological samples and 20 off-site samples

Cultural Sites Off-Site Areas P-value pH 4.87 ± 0.11 4.77 ± 0.12 0.57 Phosphorus 16.07 ± 1.80 19.25 ± 0.94 0.12 Potassium 99.93 ± 11.17 103.40 ± 16.78 0.86 Magnesium 116.79 ± 12.78 104.05 ± 9.25 0.42 Calcium 430.30 ± 54.28 364.88 ± 42.90 0.38 CEC 11.69 ± 0.53 11.43 ± 0.59 0.75 Zinc 3.32 ± 0.39 3.13 ± 0.22 0.67 Copper 1.85 ± 0.85 0.74 ± 0.05 0.20 Sulfur 14.54 ± 0.92 13.85 ± 1.34 0.66 Nitrogen 2678.21 ± 227.50 2904.44 ± 187.82 0.47 Soil color 5.64 ± 0.46 6.15 ± 0.48 0.46

A comparison of off-site soil characteristics between Fort Drum and Quantico was conducted to ascertain baseline soil fertility differences between these two sites. Quantico had significantly higher potassium, magnesium, cation exchange capacity, and nitrogen (alpha level of 0.05, Table 4-4). Soil color was also slightly darker at Quantico, however this difference was not statistically significant. Fort Drum off-site soils had significantly higher pH and phosphorus, as well as higher calcium, although differences in calcium concentrations were not statistically significant. Textures were similar among the two areas, sandy and sandy-loamy are prevalent in both (McDowell 1988, NRCS Web Soil Survey 2012). Growing season length is currently much longer in the Quantico area than in upstate New York where Fort Drum is located (NOAA 2012).

121 Although the climate during the last several centuries was cooler on average than the current climate (Gajewski 1988, Mayewski et al. 2004), growing season would have shown the same trend between these two areas of the eastern United States, increasing the productivity of Native

American agricultural fields in Virginia versus more northern areas. The combined influence of slightly higher soil fertility and increased growing season length would have had a positive impact on the initiation and continuation of maize-based agriculture in the Quantico area.

Table 4-4. Mean values (+/- standard error) of soil parameters of interest and p-values associated with t-tests for significant difference in means between off-site areas at Fort Drum, NY and MCB

Quantico, VA.

Fort Drum MCB Quantico P-value pH 5.18 ± 0.15 4.77 ± 0.12 0.035 Phosphorus 33.67 ± 5.22 19.25 ± 0.94 0.010 Potassium 37.53 ± 3.95 103.40 ± 16.78 0.000 Magnesium 39.47 ± 6.02 104.05 ± 9.25 0.000 Calcium 411.80 ± 92.49 364.88 ± 42.90 0.643 CEC 9.21 ± 0.72 11.43 ± 0.59 0.030 Zinc 2.05 ± 0.31 3.13 ± 0.22 0.006 Copper 0.62 ± 0.05 0.74 ± 0.05 0.104 Sulfur 25.12 ± 3.22 13.85 ± 1.34 0.002 Nitrogen 1791.28 ± 188.97 2904.44 ± 187.82 0.000 Soil color 6.70 ± 0.49 6.15 ± 0.48 0.441

Regression Tree Model Results

Regression tree modeling of archaeological and off-site data from both Fort Drum and

Quantico as a combined data set revealed a strong influence of Site of Origin (i.e., which site the data originated from) as a predictor of soil characteristics (Table 4-5). Specifically, Site of Origin was shown to be a predictor of highest importance (i.e., mean square error of the model increased if Site of Origin was replaced with a randomly chosen predictor) for response variables of pH,

122 phosphorus, magnesium, sulfur, and soil color. It was the second most important predictor of calcium, cation exchange capacity, zinc, and nitrogen. Site type (archaeological or off-site) typically displayed a negative percent increase in mean square error, indicating that when Site

Type was replaced with a random variable, mean square error of the model actually decreased.

Correlation among variables was apparent as well; pH, cation exchange capacity, and calcium showed high correlation, as well as potassium and magnesium levels. Models displayed an average of 52.3% of variance explained for all response variables. This encompassed a wide range of values, from -7.5% of the variance in copper levels explained, to 79.2% of the variance in sulfur values explained. Of essential nutrients (phosphorus, potassium, magnesium, calcium, and nitrogen), the models explained a maximum of 74.4% of variance (magnesium), and a minimum of 27.3% (potassium).

Table 4-5. Percent increase in mean square error (MSE) of all predictors for each response variable (soil characteristic), lumping data from both Fort Drum, NY and Quantico, VA.

Predictors of highest importance for each response variable are indicated by bold red text.

Percent increase in Mean Square Error (MSE) % Variance Response Base Site of Origin Site Type Plot pH P K Mg Ca CEC N Soil Color explained pH 9.0 36.1 -0.4 -2.1 ** 7.6 13.0 8.5 34.2 19.4 9.7 0.9 69.1 P 3.2 31.2 3.5 -1.2 4.6 ** 7.9 10.0 5.5 13.2 14.2 8.4 67.0 K 6.1 8.9 -1.0 -2.7 11.8 -0.6 ** 18.6 6.0 6.9 6.1 1.4 27.3 Mg 10.0 27.1 2.9 0.2 5.0 4.4 25.4 ** 21.0 8.9 5.2 -0.1 74.4 Ca 7.1 24.6 -1.5 -1.9 29.1 0.6 7.5 17.7 ** 7.3 5.1 5.6 74.0 CEC 4.1 27.4 -1.0 -1.8 18.9 12.6 7.9 10.7 9.1 ** 39.0 1.8 68.2 Zn 3.2 18.9 0.6 -0.6 6.6 4.5 7.9 9.0 13.2 3.2 5.1 7.8 34.3 Cu 0.5 -3.0 0.7 -2.8 1.5 -0.2 -3.4 6.1 3.7 0.7 2.7 0.2 -7.5 S 3.2 21.0 0.8 -1.9 8.1 10.4 15.7 16.5 14.6 7.7 11.6 -0.5 79.2 N 3.2 27.2 -2.0 -0.7 3.6 10.8 19.7 7.5 8.9 33.4 ** 2.8 63.6 Soil Color 3.4 32.3 4.7 1.9 0.2 10.6 -0.2 5.0 1.6 9.0 6.1 ** 25.3

123 The variable ‘Military Base’ (equal to Base in Table 4-5) was not indicated as a strong predictor in the combined model. However, as differences in soil type exist between these two geographically removed sites, combining the data may have been masking trends that are specific to geographic region, climate, underlying soil type, or Native American land uses that vary between these areas. Eliminating Military Base as a confounding factor was achieved by running two separate sets of regression tree models, one for each dataset. When each base was considered individually, model predictions were improved for Fort Drum, and significantly reduced for

Quantico. The percentage of the variance explained by the model for any response did not exceed 57% for Quantico individual data (Table 4-7), while the percent variance explained by models of Fort Drum data often exceeded 70% (Table 4-6). At Fort Drum, Site of Origin was again the most important predictor of several response variables, including phosphorus, magnesium, cation exchange capacity, zinc, copper, sulfur, and soil color. Site Type

(archaeological or off-site) was elevated in importance over the combined model, however percent increases in mean square error did not exceed 6.2% (soil color). Site Type was a somewhat important predictor of soil color however, corroborating results of a Welch’s t-test that indicated a significant difference between soil color on archaeological sites and off-site areas at

Fort Drum.

Results of regression tree modeling of Fort Drum data indicated that inherent differences exist between the cultural sites that were surveyed, such as soil series and topography. This could have been a confounding factor in elucidating the effect of Native American land uses on modern day soils. One way to identify the effect of Site Type (archaeological site versus off-site) on response variables of soil characteristics would be to build a regression tree model using only one archaeological/off-site pairing (same Site of Origin). However, the strength of such a model is precluded here because of low sample size, due to cultural site boundary limitations. Regression tree modeling of Quantico data revealed a slightly different pattern. Site of Origin did not have

124 the same level of importance in predicting the response variables (aside from sulfur and soil color). However, the pattern of decreased importance of Site Type as a predictor held to the same trend shown by the combined model and model of Fort Drum data only. Most notable among the results of the regression tree model of Quantico data are correlations between various measured soil characteristics, such as pH and calcium levels, and magnesium and calcium levels. The absence of one strong predictor of all response variables indicated high variation among all sites, whether archaeological or off-site, and patterns were difficult to discern.

Table 4-6. Percent increase in mean square error (MSE) of all predictors for each response variable (soil characteristic) for archaeological and off-site data from Fort Drum, NY. Predictors of highest importance for each response variable are indicated by bold red text.

Percent increase in Mean Square Error (MSE) % Variance Response Site of Origin Site Type Plot pH P K Mg Ca CEC N Soil Color explained pH 22.8 -1.3 -1.3 ** 6.5 10.1 12.3 30.0 14.2 7.3 0.9 74.3 P 26.7 3.4 -1.6 4.8 ** 4.7 6.5 5.5 16.1 12.0 7.1 62.8 K 6.5 0.9 -1.4 3.3 -1.0 ** 13.0 6.9 2.9 2.1 -1.9 14.2 Mg 25.9 1.2 -2.0 8.6 4.3 16.0 ** 20.6 7.2 7.1 0.2 87.3 Ca 12.6 1.1 -2.7 15.3 3.0 6.7 25.3 ** 3.8 0.9 6.5 76.3 CEC 25.0 1.3 -0.1 10.4 17.3 8.5 6.2 6.3 ** 37.4 5.5 80.2 Zn 16.7 1.5 -3.4 15.7 5.7 3.4 5.4 8.8 4.9 3.3 14.2 43.5 Cu 20.0 -1.2 3.2 3.0 5.6 2.0 6.7 8.4 7.0 4.2 3.7 26.0 S 18.1 1.7 0.3 12.1 9.4 15.7 9.5 16.7 8.4 10.8 3.1 78.0 N 18.5 -2.0 -0.7 2.4 13.6 21.1 8.0 7.6 37.4 ** 4.5 77.7 Soil Color 16.0 6.2 -3.0 -0.1 13.1 0.8 4.4 0.4 8.1 6.9 ** 23.1

125 Table 4-7. Percent increase in mean square error (MSE) of all predictors for each response variable (soil characteristic) for archaeological and off-site data from MCB Quantico, VA.

Predictors of highest importance for each response variable are indicated by bold red text.

Percent increase in Mean Square Error (MSE) % Variance Response Site of Origin Site Type Plot pH P K Mg Ca CEC N Soil Color explained pH 4.2 -1.3 -1.1 ** 1.9 15.3 1.4 24.2 22.2 2.9 0.4 51.4 P 13.4 3.7 -2.0 -1.6 ** 8.7 1.2 1.1 -1.3 0.4 -2.2 12.7 K -1.9 0.7 -1.7 18.9 0.2 ** 5.1 0.1 1.6 1.8 1.0 15.8 Mg 5.0 1.1 -1.6 1.2 -3.6 4.7 ** 22.8 4.5 3.7 -0.8 48.6 Ca 1.8 1.7 -2.5 25.3 0.6 4.5 20.8 ** 5.2 -3.3 1.8 54.2 CEC 1.6 -0.5 -1.0 23.6 -3.5 -1.1 7.5 3.9 ** 10.2 -2.3 33.6 Zn -0.5 -0.3 -0.3 0.4 4.3 3.9 4.9 1.6 4.2 0.4 2.5 18.6 Cu 0.9 -1.5 -1.6 0.5 2.3 -0.9 5.2 2.8 1.7 2.5 0.5 -7.4 S 25.2 0.6 2.0 2.3 -2.2 -0.7 23.9 4.7 -1.7 8.2 1.7 56.6 N 10.9 -0.7 -2.8 7.6 -0.8 11.6 4.9 1.3 14.5 ** 1.7 24.0 Soil Color 22.9 0.6 -6.1 4.1 -3.0 -3.4 -0.1 -3.8 4.5 -1.5 ** 11.7

Spatial Interpolation Results

Spatial interpolation of patterns in soil characteristics at Fort Drum revealed variable patterns in pH, with more basic conditions existing to the south of the archaeological site (Figure

4-6). However the range of pH values was very small (small standard error indicated this as well), so this pattern indicated very little difference in pH across the entire area sampled.

Interpolations of spatial patterns of cation exchange capacity (CEC), calcium, magnesium, and potassium all indicated areas of higher nutrient status to the south of Camp Drum 1 (Figures 4-6 and 4-7); however only a small area indicated higher values of magnesium than baseline soil samples (i.e., most interpolated values were negative, indicated by colors of green through dark orange). Variable patterns were indicated in soil color as well, with areas of darker soil color scattered around the site and surrounding area (indicated by lower color values; Figure 4-6). The largest area of soils darker than baseline measurements was located to the southeast of the village palisade, and overlapped the area of increased cation exchange capacity, calcium, magnesium,

126 and potassium. Patterns of phosphorus showed a very different trend, with a concentrated area of increased phosphorus in the center of the archaeological site, and levels below baseline phosphorus in the area to the south of the site where increases in other nutrients occurred (Figure

4-7). The increase in calcium, magnesium, and potassium to the south of the site could indicate that human amendment of soils occurred in this area, possibly indicating that agricultural fields were located here. The differing pattern in phosphorus levels could be indicative of a different type of human modification or influence that would have an effect on soils and be occurring most often within a village versus outside of it. Possibilities are midden or hearth creation, or processing or storage of collected food or resources. Smaller inclusions of darker soil color could be indicative of more localized influences as well, which could be elucidated with detailed map of site features.

The Camp Drum 1 archaeological site excavations and feature locations were mapped and available in hardcopy from Fort Drum CRM personnel. Unfortunately, the site map was not digitized and equipment and resources were not available to do so. Therefore, comparisons between the map and spatial patterns in nutrient concentrations across the site were conducted as a visual interpretation of correlation between nutrient concentrations and features that could possibly have produced a signal in the soil. Across the Camp Drum 1 archaeological site, ten hearths, four storage pits (for grains, etc), and two middens were identified. All of these features were located within the palisade boundary (indicated by the thin black line in Figures 4-6 and 4-

7), except one of the middens, which was located to the north of the site where a beaver dam has caused flooding over the site of the large buried ash midden. This area was not sampled.

Therefore, only the areas of increased nutrients or darker soil color located within the area of the palisade would have been associated with any of these features. There was one area within the palisade that showed pH, cation exchange capacity, calcium, and potassium higher than baseline

(yellow), and the highest levels of magnesium and phosphorus that occurred (red). This area,

127 however, showed lighter soil color (red). Features that were located in this area were a historical period horse burial and the midden area. The midden area is located more to the east of where the main catchment of many of the soil nutrients occurred, however the area of increased phosphorus does encompass the area of the midden. As midden creation has been said to produce “Indian dirt” in this area, soil of darker color and higher fertility (Laurie Rush, Fort Drum CRM, personal communication), the most logical cause of the catchments of increased soil fertility in association with this Native American village site is midden creation. While correlation does not equal causation, the opposite trends of these different nutrients and their spatial relationship to a known

Native American village site is notable. Exact locations of Native American features, via the use of a GPS unit to obtain coordinates for each feature, would reduce uncertainty in this visual analysis.

128

Figure 4-6. Spatial patterns in deviation from baseline values of pH, cation exchange capacity (meq/100 g), and soil color on Camp Drum 1 and surrounding area, Fort Drum, NY. Red areas indicate high values, while green areas indicate lower values. Colors of yellow to red indicate areas where measured soil characteristics were higher than baseline values. For soil color, more negative values indicate darker colors, so green, light green, and yellow are indicative of areas of soil darker than baseline soils.

129

Figure 4-7. Spatial patterns in deviation from baseline values of calcium, magnesium, potassium, and phosphorus

(ppm) on Camp Drum 1 and surrounding area, Fort Drum, NY. Red areas indicate high values, while green areas indicate lower values. Colors of yellow to red for calcium, potassium, and phosphorus indicate areas where measured soil characteristics were higher than baseline values. For magnesium, colors from green to dark orange indicate areas where magnesium values were lower than baseline, and only red areas indicate where magnesium measured was higher than baseline.

130

Canfield Island Archaeological Site in Lycoming County, PA was chosen for soils testing to bridge the gap between the most northern site, Fort Drum, and the two southern sites, Quantico and the Cheatham Annex. This site provided another sampled area that was glaciated in the last glacial maximum (Late Wisconsinan), and allowed comparisons on a latitude gradient.

Interpolated patterns across a grid of soil samples on Canfield Island indicated higher pH, cation exchange capacity, calcium, and magnesium levels on the southeast portion of the site (Figures 4-

8 and 4-9). Phosphorus and potassium showed a differing trend, with catchments areas of higher status of these nutrients on the northwest and riverside edge of the site (Figure 4-9). Phosphorus again showed an opposite trend to that of calcium and magnesium, potentially indicating that these nutrients signal different classes of land use. Soil color, as at Fort Drum, indicated a variable pattern across the site (Figure 4-8). However, the areas of the site that showed increased potassium and phosphorus also showed darker soil colors (corresponding to lower values). Areas of decreased amounts of calcium, magnesium, potassium and phosphorus, and lighter soil color, generally occurred on the northeast side of the site, away from the riverbank. Here, as at Camp

Drum 1 on Fort Drum, calcium and magnesium may be indicative of a different land use type than phosphorus and, in this case, potassium. Knowledge of exact boundaries of site features such as longhouses, and locations of hearths, middens, and storage pits, would help to elucidate the potential impact of Native American land uses in the soil, and correlate specific land use types and activities with catchment areas of different soil nutrients.

Unfortunately, baseline soil data was not available for the Canfield Island archaeological site because of the prevalence of modern agricultural fields surrounding the site. Therefore, comparisons of the high levels of nutrients in the catchment areas noted through spatial interpolation to off-site data are not possible. However, similar patterns of soil nutrient concentrations that exist here can be compared and related to the other glaciated study site, Camp

131 Drum 1 on Fort Drum. Canfield Island is located in an area of very rich soil, the Susquehanna

River flood plain. Agriculture was very important here several centuries the pre-contact (Stewart

1994, Hart and Sidell 1996). Although this area contains very rich agricultural soils, patterns of soil fertility on this important archaeological site indicate similar trends as Camp Drum 1 (Figures

4-6 and 4-7); the latter being in an area of inherently poor, highly leached, sandy glacial outwash soils (McDowell 1988). This does not support the hypothesis that in areas of richer soils, amendments would not have been necessary to initiate agricultural production, and therefore these areas would potentially show a signal of depletion rather than amendment. However, because the location of potential agricultural field areas associated with the Canfield

Archaeological site is unclear, it is not known where exactly the signal of Native American agricultural use would appear. That fields would be to the northwest or southeast of the village site seems possible, because these areas would be along the river and therefore receive rich sediment deposits due to flooding. These sediment deposits would renew the soils continually, effectively creating highly productive, fertile soils without the addition of human amendments.

The catchment areas of higher soil fertility that are apparent from spatial interpolation across the excavated area could be indicative of Native American influence in the form of midden creation, hearth creation, or another form of land use that would affect soil, and are present in a similar pattern to that of a site in a poorer soil area, Camp Drum 1. Therefore, support or refutation of the hypothesis based on this similarity in soil fertility trends between two Native American village sites in areas of differing baseline soil fertility is not possible.

132

Figure 4-8. Spatial patterns in pH, cation exchange capacity (meq/100 g), and soil color on a

Susquehannock village site on Canfield Island, Lycoming County, PA. Blue areas indicate high values, while green areas indicate lower values. For soil color, lower values are concurrent with darker soil colors, so areas of green are indicative of darker soil. Baseline soil characteristics were not available in this area due to modern land disturbance surrounding the archaeological site.

133

Figure 4-9. Spatial patterns in calcium, magnesium, potassium, phosphorus (all in ppm) on a

Susquehannock village site on Canfield Island, Lycoming County, PA. Blue areas indicate high values, while green areas indicate lower values. Baseline soil characteristics were not available in this area due to modern land disturbance surrounding the archaeological site.

Grid sampling along the northern bank of Chopawamsic Creek at Marine Corps Base

Quantico provided a more complete look at spatial trends in soil fertility in an area of several

Native American habitation sites. The grid of soil samples here covered four different archaeological sites distinct from one another, and off-site areas between then and around the area where artifacts and features were located. Spatial patterns in soil nutrient status indicated increased pH in association with sites 1-1, 1-2, and somewhat with site 1-3 (Figure 4-10).

However, there were small catchments of increased basicity throughout the central area where

134 testing revealed a lack of archaeological resources. Calcium, magnesium, and potassium showed similar patterns, with two inclusions of higher fertility in association with two of the hamlet sites along the creek bank (Figure 4-11). Catchment areas of increased fertility of these nutrients were associated with the most eastern (1-1) and second most western site archaeological sites (1-3a).

Phosphorus, similar to patterns observed on Camp Drum 1 at Fort Drum, showed a different pattern than base cations calcium, magnesium, and potassium (Figure 4-11). Spatially interpolated patterns indicated areas of increased phosphorus associated with the most western site (1-4), and a smaller area associated with the area in between sites 1-2 and 1-3, where cultural artifacts were not found. There was also a small inclusion on the second most eastern site (1-2).

These areas all showed levels of other nutrients for the most part at or below baseline measurements (i.e, negative interpolated values). Patterns in soil color somewhat followed patterns of phosphorus (Figure 4-10). Soil color was darker in association with hamlet site 1-4 and the area to the west of 1-2 showed darker color as well. However, there was another area of darker soil color associated with site 1-1, and 1-2 indicated soils about the same color as baseline.

Although catchment areas of important nutrients and darker soil color were present in this area, not all of the recorded archaeological sites had associated areas of increased fertility.

As modern inorganic fertilizers were not used in this area, Native American legacy effects in the soil, if present, would be visible if possible here. Therefore, the pattern of increased soil fertility in the form of base cations (Ca, Mg, and K) could be indicative of Native American influences in the soil. However, as the pattern is not repeated on all archaeological sites, and because of high variability among other archaeological sites throughout the Quantico area, definitive conclusions are difficult. Although archaeologists are certain agriculture was part of the lifeways of the Native Americans in this area of Virginia (Rountree 1989), it is possible that there is no legacy of Native American amendment in the soil. This could be due to the fact that soils in this area are inherently fertile (NRCS Web Soil Survey 2012), and therefore human

135 amendment was not necessary to initiate agricultural production. Alternatively, because resources gained from waterways were so important in the diet and lifeways of Native Americans in this area, agriculture was not as integral a source of food. In this case, less land would be necessary for agriculture, and longer fallow periods would be possible, decreasing the need for intensive amendments. Agriculture also may have been conducted in areas other than along waterways, although the convenience of farming land directly adjacent to other resources would potentially suggest otherwise.

136

Figure 4-10. Spatial patterns in deviation from baseline values of pH, cation exchange capacity

(meq/100 g), and soil color on the site of several hamlet-style habitations on the north side of

Chopawamsic Creek, Quantico, VA. Red areas indicate higher values, while green areas indicate lower values. Colors of yellow, orange, and red indicate areas where measured soil characteristics displayed values higher than baseline characteristics. For soil color, lower values indicate darker colors, and negative values indicate areas where soil was darker than baseline; therefore areas of yellow to green indicate darker color.

137

Figure 4-11. Spatial patterns in deviation from baseline values of calcium, magnesium, potassium, and phosphorus on the site of several hamlet-style habitations on the north side of

Chopawamsic Creek, Quantico, VA. Red areas indicate higher values, while green areas indicate lower values. Colors of yellow, orange, and red indicate areas where measured soil characteristics displayed values higher than baseline characteristics.

Spatial patterns of soil fertility on the Cheatham Annex were displayed as graduated symbology, where larger symbols indicate increased status of pH, CEC, Munsell soil color values

(indicating lighter color), and soil nutrient levels (Figures 4-12 and 4-13). Level of pH was variable across the sampled points (Figure 4-12). Increased basicity was in evidence in the second transect from the east, as well as samples further from the creek bank in the two western transects. High levels of cation exchange capacity were indicated in the easternmost (first)

138 transect, and in samples nearest to Queen Creek in transects one, two, and three (Figure 4-12).

Darkest soils were located in the whole of transect 2 (second from east) and the northern samples of transect 4 (nearest the creek; Figure 4-12). The soils with the lightest color were located in samples further from the creek in transects 1, 3, and 4, although transects 1 and 3 indicated lighter soil colors overall. Calcium and magnesium showed increased levels in the northernmost sample of the first transect (nearest to the creek bank; Figure 4-13). Magnesium and potassium also increased in individual samples of the second and third transects from the east. Potassium continued to show this trend in the fourth transect, indicating increased nutrient status in samples furthest from the creek bank in this transect (Figure 4-13). Phosphorus showed a spatial pattern similar to that of magnesium (Figure 4-13).

On the Cheatham Annex, unlike Fort Drum, Quantico, and Canfield Island, there is no evidence of long-term habitation by Native Americans in the form of ethnohistorical evidence or archaeological findings of artifacts identified with a village site (Bruce Larson, NAVFAC Mid

Atlantic Cultural Resources, personal communication, see Chapter 2, figure 2-11; Virtual

Jamestown 2000). Thus, we would not expect signals in soil related to agriculture. However, use of the area by Native Americans is evident in the form of large areas of shell middens in several locations along the south bank of Queen Creek. The locations of these middens are in the area of the soil sample transects. Therefore, it was hypothesized that areas in closer proximity to the creek bank would potentially indicate signals in the soil from inadvertent amendment, such as with midden creation. In this area, the main occupation of the Native Americans was collecting shellfish and fish from Queen Creek (Bruce Larson, NAVFAC Mid Atlantic Cultural Resources, personal communication, and personal observation of shell midden evidence). Shell middens may return much higher calcium than off-site soils because freshwater mussel shells are composed mainly of calcium carbonate. In the analysis conducted here, the northernmost sample of the easternmost transects have very high calcium measurements. These samples may have

139 been taken in an area significantly affected by shellfish midden creation, which would be indicative of Native American activity. In addition, samples nearest the creek bank across the four transects indicated higher values of cation exchange capacity, and in some transects, increased calcium and magnesium were found in samples nearest the creek bank. However, transect four (most western transect) indicated higher values of cation exchange capacity, calcium, magnesium, and potassium in sample furthest from the creek bank, which did not support the hypothesis.

Figure 4-12. Spatial patterns of pH, cation exchange capacity (meq/100 g), and soil color on the south bank of Queen Creek in the Wilderness Area, Cheatham Annex, VA. Larger symbols indicate samples with higher measured values of soil parameters of interest. For soil color, smaller values indicate darker colors, therefore smaller symbols are indicative of areas of darker soils.

140

Figure 4-13. Spatial patterns of calcium, magnesium, potassium, and phosphorus (all in ppm) on the south bank of Queen Creek in the Wilderness Area, Cheatham Annex, VA. Larger symbols indicate samples with higher measured values of soil parameters of interest.

Lumping archaeological site data together across multiple locations was shown by regression tree modeling to be an insufficient methodology in elucidating the overall effect of

Native Americans on present-day soils. Consideration of archaeological sites as separate entities therefore may be necessary. Spatial interpolation of patterns in soil characteristics is a useful way of characterizing one single relatively intensively inhabited site, such as a village. Interpretation of spatial patterns in soil characteristics was not limited by averaging together data from all sample points, and lumping across disparate sites is avoided. Although statistical analysis is difficult to conduct due to problems of pseudoreplication and spatial correlation, the spatial

141 distribution of areas of increased nutrient status in relation to cultural resources is a feasible way to visualize changes in soil characteristics due to the presence or absence of Native American land use legacies.

Comparisons between means and ranges of soil pH, cation exchange capacity, soil color, and nutrient concentrations in the four sites were made using soil test results from grid samples

(Table 4-8). All of these samples were taken on and around significant archaeological sites. As the soil samples were taken in the same condition (i.e., in culturally significant areas), comparisons were possible. Because the samples are not lumped in this analysis, the full range of values is apparent at each site. This allowed comparisons of maximum and minimum values, which were indicative of catchment areas of increased fertility, and for that reason, are integral to understanding differences between the study sites. Generally, Drum indicated the lowest fertility, except in phosphorus and maximum calcium measurements (Table 4-8). Canfield Island showed the highest mean magnesium and calcium, while Quantico showed the highest mean potassium and cation exchange capacity, and the highest maximum concentration of magnesium. Canfield

Island also showed the darkest mean soil color among samples, and darkest maximum soil color, indicating elevated levels of soil organic matter. The Cheatham Annex had the highest maximum calcium measured among the four sites, which could be due to high inputs of shells to the soils during Native American shell midden creation. Single soil samples that indicate high levels of nutrients (such as calcium at Cheatham Annex), while the mean values are much lower, could be indicative of a “point-source” of Native American amendments in soil. This is in contrast to a more “broadcast” source, such as would be apparent with a signal of agricultural amendments.

142 Table 4-8. Mean values (+/- standard error) and ranges of values of soil parameters of interest for fertility comparisons between grids of soil samples taken at significant archaeological sites at Fort

Drum, NY, Canfield Island, PA, MCB Quantico, VA, and the Cheatham Annex, VA.

Fort Drum Canfield Island MCB Quantico Cheatham Annex mean ± SE range mean ± SE range mean ± SE range mean ± SE range pH 5.8 ± 0.07 5.2 - 7.4 6.1 ± 0.03 5.8 - 6.6 5.0 ± 0.07 3.8 - 6.5 6.1 ± 0.13 4.9 - 7.4 P 52.9 ± 7.25 3.0 - 291.0 14.9 ± 1.04 7.0 - 32.0 14.1 ± 0.65 5.0 - 35.0 26.4 ± 2.77 6.0 - 51.0 K 39.0 ± 2.98 17.0 - 107.0 71.1 ± 4.87 34.0 - 148.0 96.8 ± 4.21 34.0 - 229.0 63.1 ± 4.04 35.0 - 108.0 Mg 29.3 ± 1.63 14.0 - 59.0 200.3 ± 9.47 124.0 - 299.0 131.5 ± 8.46 24.0 - 411.0 92.5 ± 5.82 41.0 - 137.0 Ca 606.2 ± 90.50 153.0 - 3540.0 1167.8 ± 43.78 797.0 - 1998.0 414.5 ± 34.71 68.0 - 2152.0 1088.3 ± 144.93 239.0 - 3553.0 CEC 7.1 ± 0.34 3.9 - 16.8 11.2 ± 0.24 9.0 - 15.3 12.0 ± 0.31 7.5 - 21.3 9.2 ± 0.46 5.7 - 16.2 Color 5.7 ± 0.29 2 - 12 4.7 ± 0.19 2 - 6 7.0 ± 0.25 2 - 14 8.3 ± 0.84 3 - 14

Conclusion

Patterns of soil fertility on archaeological sites were investigated to elucidate the effects of Native American influences at the study sites. This included areas of inherently high versus low fertility soils. The findings of these studies suggest that Native Americans may have had an influence in areas with intensive habitation. For example, at Fort Drum archaeological sites presently have soils that display darker color and higher over-all fertility. It is interesting to note that Fort Drum exists on inherently infertile glacial outwash that would require soil amendments to foster agriculture. The three other study sites (Quantico, Canfield Island, and the Cheatham

Annex) are along waterways. The results of this study suggest that Native American influences were less discernible at these locations. This may be due to their inherently more fertile soils, which would not require Native American amendments to facilitate agriculture. Or, in the case of the Cheatham Annex, gathering of important resources such as shellfish from the brackish waterways of the tidal area of Virginia could have been the main activity of Natives in the area, negating the need for agriculture.

The results of these analyses suggest that patterns are site-specific as well as region- specific. The strong influence of Site of Origin on the vast majority of soil characteristics

143 suggests that intensive sampling of individual archaeological sites, in comparison with associated off-site areas, is necessary. This would eliminate the need to lump data across archaeological sites that show inherent differences in geology and topography, leading to landscape heterogeneity in soil characteristics. Spatial interpolation of patterns in soil characteristics across a single archaeological site was used successfully to locate catchments of increased fertility, indicating that sampling soils in a grid across an archaeological site and surrounding area is potentially more effective than lumping on- versus off-site data because catchments or inclusions of increased soil fertility are more evident. Further information, including baseline soil fertility characteristics and detailed maps of feature locations in intensively inhabited sites, is necessary to determine the level of correlation between specific Native American land uses and soil fertility characteristics.

144 Literature Cited

Abrams, M.D., and Nowacki, G.J. 2008. Native Americans as active and passive promoters of

mast and fruit trees in the eastern USA. The Holocene 18: 1123-1137.

Bamann, S., Kuhn, R., Molnar, J. and Snow, D. 1992. Iroquoian Archaeology. Annual Review of

Anthropology 21: 435-460.

Black, B.A., and Abrams, M.D. 2001. Influences of Native Americans and surveyor biases on

metes and bounds witness-tree distribution. Ecology 82: 2574-2586.

Blume, H. and Leinweber, P. 2004. Plaggen soils: Landscape history, properties, and

classification. Journal of Plant Nutrition and Soil Science 167: 319-327.

Bokhorst, M.P., Duller, G.A.T., and Van Mourik, J.M. 2005. Optical dating of a Fimic Anthrosol

in the southern Netherlands. Journal of Archaeological Science 32: 547-553.

Breiman, L. 2001. Random Forests. Machine Learning 45: 5-32.

Butzer, K.W. 1990. The Indian legacy in the American landscape. In The Making of the American

Landscape. Conzen, M.P., editor. Unwin and Hyman, Boston, Massachusetts. Pages 27-

50.

Ceci, L. 1990. Squanto and the Pilgrims. Society 27: 40-44.

Ceci, L. 1975. Fish fertilizer: A native North American practice? Science 4: 26-30.

Day, G.M. 1953. The Indian as an ecological factor in the northeastern forest. Ecology 34: 329-

346.

Delcourt, H.R. 1987. The impact of prehistoric agriculture and land occupation on natural

vegetation. TREE 2: 39-44.

Doolittle, W.E. 2004. Permanent vs. shifting cultivation in the Eastern Woodlands of North

America prior to European contact. Agriculture and Human Values 21: 181-189.

145 Doolittle, W.E. 1992. Agriculture in North America on the eve of Contact: A reassessment.

Annals of the Association of American Geographers 82: 386-401.

Dudal, R., Nachtergaele, F., and Purnell, M.F. 2002. The human factor of soil formation. 17th

World Congress of Soil Science, August 14-21, Thailand. Symposium no. 18, Paper no.

93.

ESRI 2010. ArcGIS Desktop: Release 10. Redlands, California: Environmental Systems Research

Institute.

Franzmeier, D.P. 1988. Relation of organic matter content to texture and color of Indiana soils.

Proceedings of the Indiana Academy of Science 98: 463-471.

Fritz, G.J. 1990. Multiple pathways to farming in precontact eastern North America. Journal of

World Prehistory 4: 387-435.

Gajewski, K. 1988. Late Holocene climate changes in eastern North America estimated from

pollen data. Quaternary Research 29: 255-262.

Hart, J.P. and Sidell, N.A. 1996. Prehistoric agricultural systems in the West Branch of the

Susquehanna River Basin, A.D. 800 to A.D. 1350. Northeast Anthropology 52: 1-30.

Hecht, S.B. 2004. Indigenous soil management and the creation of Amazonian dark earths:

Implications of Kayapó practices. In Amazonian Dark Earths: Origin, Properties,

Management. Lehmann, J., Kern, D.C., Glaser, B., and Woods, W.I., editors. Kluwer

Academic Publishers, New York, New York. Pages 355-372.

Hesse, R. and Baade, J. 2009. Irrigation agriculture and the sedimentary record in the Palpa

Valley, southern Peru. Catena 77: 119-129.

Hurt, R.D. 2002. American Agriculture: A Brief History. Purdue University Press, West

Lafayette, Indiana.

International Plant Nutrition Institute (IPNI). 2010. Soil pH and the availability of plant nutrients.

Plant Nutrition TODAY No. 2.

146 Jácome, A.G. 1993. Management of land, water and vegetation in traditional agro-ecosystems in

Central Mexico. Landscape and Urban Planning 27: 141-150.

Jiménez-Osornio, J.J. and Gomez-Pompa, A. 1991. Human role in shaping of the flora in a

wetland community, the chinampa. Landscape and Urban Planning 20: 47-51.

Kögel-Knabner, I., Amelung, W., Cao, Z., Fiedler, S., Frenzel, P., Jahn, R., Kalbitz, K., Kölbl,

A., and Schloter, M. 2010. Biogeochemistry of paddy soils. Geoderma 157: 1-14.

Lehmann, J., Pereira da Silva Jr., J., Steiner, C., Nehls, T., Zech, W., and Glaser, B. 2003.

Nutrient availability and leaching in an archaeological Anthrosol and a Ferralsol of the

Central Amazon basin: Fertilizer, manure and charcoal amendments. Plant and Soil 249:

343-357.

Lewandowski, S. 1987. Diohe’ko, the Three Sisters in Seneca life: Implications for a native

agriculture in the Finger Lakes region of New York State. Agriculture and Human Values

4: 76-93.

Liaw, A. and Wiener, M. 2002. Classification and regression by randomForest. R News 2: 18-22.

Loeb, R.E. 1998. Evidence of prehistoric corn (Zea mays) and hickory (Carya spp.) planting in

New York City: Vegetation history of Hunter Island, Bronx County, New York. Journal

of the Torrey Botanical Society 125: 74-86.

Mann, C.C. 2000. The good earth: Did people improve the Amazon Basin? Science 4: 788-790.

Mayewski, P.A., Rohling, E.E., Stager, J.C., Karlén, W., Maasch, K.A., Meeker, L.D., Meyerson,

E.A., Gasse, F., van Kreveld, S., Holmgren, K., Lee-Thorp, J., Rosqvist, G., Rack, F.,

Staubwasser, M., Schneider, R.R., and Steig, E.J. 2004. Holocene climate variability.

Quaternary Research 62: 243-255.

McDowell, L. 1988. Soil survey of Jefferson County, New York. USDA Soil Conservation

Service.

147 McFadgen, B.G. 1980. Maori plaggen soils in New Zealand, their origin and properties. Journal

of the Royal Society of New Zealand 10: 3-18.

Mengel, K. and Kirkby, E.A. 2001. Principles of Plant Nutrition. Springer-Science + Business

Media, Dordrecht.

Minnesota Department of Natural Resources (MN DNR). 2008. DNRGarmin 5.4.1 application.

http://www.dnr.state.mn.us/mis/gis/tools/arcview/extensions/DNRGarmin/DNRGarmin.html

Mrozowski, S.A. 1994. The discovery of a Native American cornfield on Cape Cod. Archaeology

of Eastern North America 22: 47-62.

Mücher, H.J., Slotboom, R.T., and ten Veen, W.J. 1990. Palynology and micromorphology of a

man-made soil: A reconstruction of the agricultural history since late-Medieval time of

the Posteles in the Netherlands. Catena 17: 55-67.

Munsell Color Company. 1975. Munsell soil color charts, 1975 edition. Munsell Color Company,

Baltimore, Maryland.

National Oceanic and Atmospheric Administration (NOAA), National Climatic Data Center.

Climate Data Online website accessed November 2012.

http://www.ncdc.noaa.gov/cdo-wed

Natural Resources Conservation Service (NRCS), Soil Survey Staff. 2012. Web Soil Survey.

United States Department of Agriculture.

http://websoilsurvey.nrcs.usda.gov/

Pape, J.C. 1970. Plaggen soils in the Netherlands. Geoderma 4: 229-255.

Pearson, C.S. and Cline, M.G. 1961. Soil Survey, Lewis County, New York. USDA Soil

Conservation Service.

R Core Team. 2012. R: A Language and Environment for Statistical Computing. R Foundation

for Statistical Computing, Vienna, Austria.

http://www.R-project.org

148 Rountree, H.C. 1989. The Powhatan Indians of Virginia: Their Traditional Culture. University of

Oklahoma Press, Norman, Oklahoma.

Skinner, S.M. 1986. Phosphorous as an anthrosol indicator. Midcontinental Journal of

Archaeology 11: 51-78.

Smith, B.D. 1989. Origins of agriculture in eastern North America. Science 246: 1566-1571.

Steiner, C., Teixeira, W.G., Lehmann, J., Nehls, T., Vasconcelos de Macêdo, J.L., Blum, W.E.H.,

and Zech, W. 2007. Long term effects of manure, charcoal and mineral fertilization on

crop production and fertility on a highly weathered Central Amazonian upland soil. Plant

and Soil 291: 275-290.

Stewart, M. 1993. Comparison of Late Woodland cultures: Delaware, Potomac, and Susquehanna

River valleys, Middle Atlantic region. Archaeology of Eastern New York 21: 163-178.

Stewart, R.M. 1994. Prehistoric Farmers of the Susquehanna Valley – Clemson Island Culture

and the St. Anthony Site. Occasional Publications in Northeastern Anthropology, No. 13.

Moeller, R.W., general editor. Archaeological Services, Bethlehem, Connecticut.

Struever, S. and Vickery, K.D. 1973. The beginnings of cultivation in the Midwest-riverine area

of the United States. American Anthropologist 75: 1197-1220.

Sykes, C.M. 1980. Swidden horticulture and Iroquoian settlement. Archaeology of Eastern North

America 8: 45-52.

Trigger, B.G. 1978. Handbook of North American Indians – Northeast, Volume 15. Sturtevant,

W.C., editor. Smithsonian Institution, Washington, D.C.

Virtual Jamestown. 2000. John Smith’s Map of Virginia, 1624 Rectified: Chesapeake Bay.

Virginia Center of Digital History, University of Virginia.

http://www.virtualjamestown.org/johnsmithsmap.html

149 Woods, W.I. 1984. Soil chemical investigations in Illinois archaeology: Two example studies. In

Archaeological Chemistry – III. Lambert, J.B., editor. American Chemical Society,

Washington, D.C.

Chapter 5

Conclusion

A protocol for efficient identification and protection of cultural resources

Present day vegetation and soils associated with archaeological sites overall show trends that differ among the four sites studied here. Geography influences settlement patterns, available resources, and baseline soil fertility (Black and Abrams 2001a, Black and Abrams 2001b). For example, Native Americans in more northern regions seem to have relied more heavily on hunting and gathering and seasonal movement to resource camps, possibly because of the later introduction of cold-hardy maize varieties (Stothers and Yarnell 1977, Pagoulatos 1992, Abrams and Nowacki 2008). Settlement patterns in this area show more regular distribution across the landscape, not necessarily tied to major waterways. At the same time, groups living in warmer, more amenable environments further south seem to have gravitated towards waterways, initially because of the resources and transportation they provided, but also because they provided very fertile ground for growing neotropical crops (Stewart 1993, Doolittle 2004). Application of the methodologies described in this dissertation therefore requires careful examination of geographic variables and anthropological context. I suggest that trends would be apparent on an EPA Level

III ecoregional basis (Figure 5-1). The Level III ecoregions are derived from Omernik (1987), and designed to serve as a spatial framework for environmental resource management. They denote areas within which the type of ecosystems and environmental resources are generally similar. In this particular study, sites in the Eastern Great Lakes Lowlands region (region #83;

Fort Drum), the Ridge and Valley region (region #67; Canfield Island), and the Southeastern

Plains region (region #65; MCB Quantico and the Cheatham Annex) were considered.

151

Figure 5-1. EPA Level III Ecoregions of the eastern United States.

152 Cultural Resources Management (CRM) personnel employed by the United States government are mandated to protect prehistoric and historic resources of cultural importance on military installations throughout the United States (DoD Legacy Program, https://www.dodlegacy.org/legacy/index.aspx). On vast land areas owned by the US Military, this mandate necessitates efficient identification and characterization of cultural resources to ensure they are protected, while still allowing military operations to occur. Predictive modeling techniques to streamline the process of site identification are currently utilized by CRM personnel, employing geographic variables such as topography and elevation (Zeidler et al.

2001). These models do provide predictive capability, but could be improved through addition of other important landscape variables, such as vegetation and soils, that would allow targeted areas to be narrowed and prioritized more efficiently. This research provides a more complete understanding of Native American land use legacies on the present day landscape in the vegetation and soils, and highlights important indicator species and soil fertility trends at different locations throughout the eastern United States.

Important indicator species of vegetation were identified at Fort Drum and MCB

Quantico. Indicator species were able to be identified at the Cheatham Annex; however the ubiquitous and possibly low-level use of the area made catchments of these species associated with archaeological sites difficult to discern. At Fort Drum a very important and indicative vegetation association was identified, the oak-pine-blueberry (Quercus-Pinus-Vaccinium) association of the sandy glacial outwash moraine. At MCB Quantico, white oak (Q. alba L.) was shown to be an important indicator species of cultural resources. At both of these installations, archaeological sites were shown to be distinctly more open (decreased trees per acre) than off-site plots. Soil charcoal, while more characteristic of archaeological sites than off-site areas, was fairly abundant across all surveyed areas, indicating the influence of Native American broadcast burning in present-day soils (e.g., Patterson and Sassaman 1988, Whitney 1994, Declourt and

153 Delcourt 1997, Williams 2002). Using vegetative indicator species for efficient identification of cultural sites requires ethnobotanical literature review to determine which species were highly important in the diet and life of Native American groups of interest. These species must be very significant, and the plant parts utilized (and thus gathered, cached, and transported) must be the seed and/or seed coat, to have facilitated the movement and subsequent increase in importance of these species in areas of Native American habitation. Also of potential importance are generational longevity and stability, making species such as oaks and hickories possibly more useful as indicators than short-lived herbaceous species. A general inventory of vegetation should be performed in an area of interest, paying close attention to topography and geology (data that can easily be gathered and displayed using GIS). Catchments of indicator species indicate landscapes or areas where the likelihood of cultural significance is high. The distribution of the indicator species may adhere closely to a certain topographic position, which could provide even more information about where archaeological sites could potentially be located, if these topographic positions can be mapped across a wider landscape.

The findings presented here suggest that a legacy of Native American agriculture in the soils of eastern North America is present. This influence may be more prevalent and thus more apparent in areas of intensive habitation and inherently low soil fertility that would require soil amendments to foster agriculture. In these areas, the amount of labor required to periodically move fields as productivity declined would have made it more necessary to take action to extend the productive life of current fields (Doolittle 1992, Abrams and Nowacki 2008). Soil amendments such as charcoal, human waste, and organic refuse would have darkened and increased the fertility of soils (Delcourt 1987, Trigger 1978, Steiner et al. 2007), conditions that may exist to the present day. Amending soil may not have happened along fresh waterways

(where soils are usually more fertile; see Stewart 1993 and Chapter 4 of this document, table 4-8), or in areas where agriculture wasn’t the main source of food (see Fritz 1990, personal

154 communications from CRM personnel at study sites). These areas include sites of brackish water, as in the Cheatham Annex, or areas further from the core of Mississippian cultural traits, as in northern Iroquoia at Fort Drum. High variation among sites surveyed (even at the same military installation) suggests that the sampling methodology to determine soil fertility should be intensive across the area of interest, as was performed in this study on several areas of known long-term habitation (Camp Drum 1 at Fort Drum, north shore of Chopawamsic Creek at MCB Quantico, and the Canfield Island Site in Lycoming County, Pennsylvania). However, because intensive sampling was needed to discern soil fertility differences associated with Native American habitation, this method would be less useful to quickly and efficiently evaluate archaeological significance. Nevertheless, this methodology provided a very useful means of elucidating settlement patterns and lifeways in various locations already known as highly cultural significant.

For example, this methodology could assist Cultural Resources Managers to locate middens and agricultural fields through catchments of increased soil fertility and darker color, two key archaeological components for interpretation of Native American village dynamics (Boone 1987,

Gremillion 1998, Mrozowski 1994).

The sites used in this study have very different geological, vegetation, prehistoric and historical characteristics. Yet similarities in Native American land uses, such as agriculture, land clearing, the use of fire and the importance of mast-bearing tree species in the diet (Abrams and

Nowacki 2008) allowed for the application of similar sampling methodology and similar ways of making conclusions. The conclusions of this study provide the framework for development of a

Cultural Resources Management Decision Support Tool. This provides planning guidance according to pre-specified information inputs, such as ecoregion, baseline soil fertility value ranges (e.g., baseline soil phosphorous range, in ppm), and dominant forest type (Table 5-1).

Within a Decision Support framework, the model inputs would also include the major food production methods of the Native Americans of the Late Woodland and/or Mississippian Time

155 Period (i.e., shellfish and/or fish, maize (Zea mays L.) based agriculture, hunting and gathering, or some combination of these methods), post-European land use (presence of modern agricultural techniques, etc), geotopographical variables such as latitude, longitude, and elevation, and forest measurements such as trees per hectare and basal area per hectare (Table 5-1). Ethnobotanical literature review, already completed for the three ecoregions included here (Appendix), would be crucial as well for CRM personnel employing this Tool. Using the results of this study, in conjunction with further studies of heavily inhabited areas and other ecoregions, a CRM Decision

Support Tool could be developed that would assist cultural resources managers in identifying relatively constrained areas of potential archaeological sites (Figure 5-2).

Table 5-1. Essential factors of the research protocol employed in this study, categorized by the data collection method. To initially assess indicator vegetation and soil fertility characteristics of archaeological sites, data from all three columns should be collected. For CRM personnel wishing to employ a Decision Support Tool utilizing this information, data collection would be limited to column three, fieldwork.

Background data collection Geographic data (GIS) Fieldwork

Determine Ecoregion Latitude, Longitude Collect and test soil samples

Ethnobotanical review (indicator species) Elevation Survey vegetation (overstory and understory)

Native Americans present Topography Determine trees/ha & basal area/ha

Lifestyles: farming, fishing, hunting & gathering Slope Post-European land usage Soil series & baseline fertility

Dominant forest type

156

Figure 5-2. Required inputs for development and use of a CRM Decision Support Tool. Boxes at the top highlight information required for fully developing the DST, and boxes at the bottom include information that would be required from a user of the DST.

This research defined two major vegetation indicators, white oak at MCB Quantico and the sand plain association at Fort Drum, as well as increases in soil fertility on archaeological sites, especially in areas of low baseline fertility such as at Fort Drum. This work lays the foundation for further studies to advance our knowledge of the driving factors in vegetation distribution and soil characteristics across archaeological landscapes, for both cultural and natural resources management personnel. Future studies should more fully evaluate the relationship between these trends and fine-scale differences in soil type, including exploration of regionally relevant quantitative baseline soil data. A focus on more heavily inhabited village sites to apply

157 these methods would help to further elucidate the relationship between long-term, intensive habitation and soil and vegetation characteristics and impacts. Determining indicator species for other ecoregions (Figure 5-1) would also aid cultural resources managers in areas to which these results don’t easily translate. Assessments in other ecoregions should utilize ethnohistorical accounts, witness tree data, fire and logging histories, bedrock and soils information, and post-

European land use information. These methods have wide application in the field of cultural resources management, and with slight modifications, could be applied anywhere in North

America.

158 Literature Cited

Abrams, M.D. and Nowacki, G.J. 2008. Native Americans as active and passive promoters of

mast and fruit trees in the eastern USA. The Holocene 18: 1123-1137. a. Black, B.A. and Abrams, M.D. 2001. Influences of Native Americans and surveyor biases on

metes and bounds witness-tree distribution. Ecology 82: 2574-2586. b. Black, B.A. and Abrams, M.D. 2001. Analysis of temporal variation and species-site

relationships of witness tree data in southeastern Pennsylvania. Canadian Journal of

Forest Research 31: 419-429.

Boone, J.L., III. 1987. Defining and measuring midden catchment. American Antiquity 52: 336-

345.

Delcourt, H.R. 1987. The impact of prehistoric agriculture and land occupation on natural

vegetation. TREE 2: 39-44.

Delcourt, H.R. and Delcourt, P.A. 1997. Pre-Columbian Native American use of fire on southern

Appalachian landscapes. Conservation Biology 11: 1010-1014.

Doolittle, W.E. 2004. Permanent vs. shifting cultivation in the Eastern Woodlands of North

America prior to European contact. Agriculture and Human Values 21: 181-189.

Doolittle, W.E. 1992. Agriculture in North America on the eve of Contact: A reassessment.

Annals of the Association of American Geographers 82: 386-401.

Fritz, G.J. 1990. Multiple pathways to farming in precontact eastern North America. Journal of

World Prehistory 4: 387-435.

Gremillion, K.J. 1998. Changing roles of wild and cultivated plant resources among early farmers

of eastern Kentucky. Southeastern Archaeology 17: 140-157.

Mrozowski, S.A. 1994. The discovery of a Native American cornfield on Cape Cod. Archaeology

of Eastern North America 22: 47-62.

159 Omernik, J.M. 1987. Ecoregions of the conterminous United States. Map (scale 1:7,500,000).

Annals of the Association of American Geographers 77: 118-125.

Pagoulatos, P. 1992. Native American land use patterns of New Jersey. Journal of Middle

Atlantic Archaeology 8: 49-65.

Patterson, W.A. and Sassaman, K.E. 1988. Indian fires in the prehistory of New England. In

Holocene Human Ecology in Northeastern North America. Nicholas, G.P., editor.

Plenum Press, New York. Pages 107-135.

Steiner, C., Teixeira, W.G., Lehmann, J., Nehls, T., Vasconcelos de Macêdo, J.L., Blum, W.E.H.,

and Zech, W. 2007. Long term effects of manure, charcoal and mineral fertilization on

crop production and fertility on a highly weathered Central Amazonian upland soil. Plant

and Soil 291: 275-290.

Stewart, M. 1993. Comparison of Late Woodland cultures: Delaware, Potomac, and Susquehanna

River valleys, Middle Atlantic region. Archaeology of Eastern New York 21: 163-178.

Stothers, D.M. and Yarnell, R.A. 1977. An agricultural revolution in the Lower Great Lakes. In

Geobotany VIII. Romans, R.C., editor. Springer US, New York City, New York. Pages

209-232.

Trigger, B.G. 1978. Handbook of North American Indians – Northeast, Volume 15. Sturtevant,

W.C., editor. Smithsonian Institution, Washington, D.C.

Whitney, G.G. 1994. From coastal wilderness to fruited plain: A history of environmental change

in temperate North America from 1500 to the present. Cambridge University Press, New

York City, New York.

Williams, G.W. 2002. Aboriginal use of fire: Are there any “natural” plant communities? In

Wilderness and Political Ecology: Aboriginal Land Management – Myths and Reality.

Kay, C.E. and Simmons, R.T., editors. University of Utah Press, Logan, Utah.

160 Zeidler, J.A. 2001. Dynamic modeling of landscape evolution and archaeological site

distributions: A three-dimensional approach. SERDP Project CS-1130. Center for

Environmental Management of Military Lands TPS 01-8, Colorado State University, Fort

Collins, Colorado.

Appendix

Ethnobotanical Inventory & Data Sources

A. Ethnobotanical information gathered through personal communication

New York

Ethnobotanical information was gathered for northern New York through personal communication with Laurie Rush, Meg Schulz, Butch Schulz, and others with Fort Drum Cultural

Resources Management.

Common Name Scientific Name Use Food: Mast White oak Quercus alba Acorns boiled and ground for flour Swamp white oak Q. bicolor Acorns boiled and ground for flour Northern red oak Q. rubra Acorns boiled and ground for flour Hickory Carya spp. Nuts collected, dried, and eaten raw American beech Fagus grandifolia Nuts collected and eaten fresh Food: Fruit Black cherry Prunus serotina Fruit eaten fresh or dried Serviceberry Amelanchier arborea Fruit eaten fresh or dried Blueberry Vaccinium angustifolium & V. corymbosum Fruit eaten fresh, dried, or cooked Black huckleberry Gaylussacia baccata Ripe fruits eaten fresh or dried Blackberry/raspberry Rubus spp. Fruit eaten fresh or dried Wild strawberry Fragaria virginiana Ripe fruits eaten fresh Mayapple Podophyllum peltatum Ripe fruits eaten fresh (also used medicinally) Food: Tuber Roots dug in fall and eaten fresh (also used Indian cucumber root Medeola virginiana medicinally) Broadleaf arrowhead (duck potato ) Sagittaria latifolia Roots, stems, and leaves processed, boiled, and eaten Arrow arum (tuckahoe) Peltandra virginica Roots, stems, and leaves processed, boiled, and eaten Food: Other Sugar maple Acer saccharum Sap collected in spring and reduced to make syrup Leaves boiled and eaten; grains collected, dried, and Lamb's-quarters Chenopodium album ground Violet Viola spp. Leaves and stems boiled and eaten as spring green Medicinal Sweet fern Comptonia peregrina Infusion of leaves used for various ailments Blue cohosh Caulophyllum thalictroides Roots used to make decoction for various ailments Eastern teaberry Gaultheria procumbens Whole plant used to make tea for stomach ailments Materials Elm Ulmus americana and U. rubra Bark used as house siding and roofing White pine Pinus strobus Stems used for dugout canoes (also used ceremonially) Stems used for dugout canoes and resin used as Red pine P. resinosa waterproofing agent Pitch pine P. rigida Pitch (sap) used as waterproofing agent Gray birch Betula populifolia Bark used for canoes, baskets, trays, etc Paper birch B. papyrifera Bark used for canoes, baskets, trays, etc Leaves used as moccasin lining (also used medicinally Common mullein (lamb’s ear ) Verbascum thapsus and ceremonially)

162 Pennsylvania

Ethnobotanical information was gathered for central and northeastern Pennsylvania through personal communication with staff of the Lycoming County Historical Society, and through literature review (see section B of this Appendix).

Common Name Scientific Name Use Food: Mast American chestnut Castanea dentata Nuts crushed or ground for meal or flour White oak Quercus alba Acorns boiled and ground for flour Swamp white oak Q. bicolor Acorns boiled and ground for flour Bur oak Q. macrocarpa Acorns boiled and ground for flour Chestnut oak Q. montana Acorns boiled and ground for flour Northern red oak Q. rubra Acorns boiled and ground for flour Black oak Q. velutina Acorns boiled and ground for flour Scarlet oak Q. coccinea Acorns boiled and ground for flour Hickory Carya spp. Nuts collected, dried, and eaten raw Black walnut Juglans nigra Nuts collected and eaten fresh or dried Butternut Juglans cinerea Nuts collected and eaten fresh or dried Food: Fruit Black cherry Prunus serotina Fruit eaten fresh or dried Serviceberry Amelanchier arborea Fruit eaten fresh or dried Elderberry Sambucus nigra Fruit eaten fresh or cooked Blueberry Vaccinium angustifolium & V. corymbosum Fruit eaten fresh, dried, or cooked Blackberry/raspberry Rubus spp. Fruit eaten fresh or dried Food: Tuber Roots dug in fall and eaten fresh (also used Indian cucumber root Medeola virginiana medicinally) Medicinal Whole plant or root used for variety of Jack-in-the-pulpit Arisaema triphyllum ailments (root also cooked to remove poison and used as food) Materials Elm Ulmus americana and U. rubra Bark used as house siding and roofing Blackgum Nyssa sylvatica Wood used for handles and teeth cleaning Bloodroot Sanguinaria canadensis Red dye for clothing and skin

163 Virginia

Ethnobotanical information was gathered for the Virginia Tidewater and coastal plain area through personal communication with Bruce Larson (NAVFAC Mid Atlantic Archaeologist),

John Haynes (former Cultural Resources Manager for Marine Corps Base Quantico), and Dave

Eckard (southern coastal Virginia local ethnobotanical authority).

Common Name Scientific Name Use Food: Mast White oak Quercus alba Acorns boiled and ground for flour Chestnut oak Q. montana Acorns boiled and ground for flour Post oak Q. stellata Acorns boiled and ground for flour Northern red oak Q. rubra Acorns boiled and ground for flour Southern red oak Q. falcata Acorns boiled and ground for flour Black oak Q. velutina Acorns boiled and ground for flour Scarlet oak Q.coccinea Acorns boiled and ground for flour Hickory Carya spp. Nuts collected, dried, and eaten raw Black walnut Juglans nigra Nuts collected and eaten fresh or dried Butternut Juglans cinerea Nuts collected and eaten fresh or dried American beech Fagus grandifolia Nuts collected and eaten fresh Food: Fruit Black cherry Prunus serotina Fruit eaten fresh or dried Chickasaw plum Prunus angustifolia Fruit eaten fresh Serviceberry Amelanchier arborea Fruit eaten fresh or dried Paw paw Asimina triloba Fruit eaten fresh or dried Persimmon Diospyros virginianan Fruit eaten fresh or cooked (also used medicinally) Elderberry Sambucus nigra Fruit eaten fresh or cooked Blueberry Vaccinium angustifolium & V. corymbosum Fruit eaten fresh, dried, or cooked Blackberry/raspberry Rubus spp. Fruit eaten fresh or dried Muscadine grape Vitis rotundifolia Fruit eaten fresh or used to make a beverage Food: Tuber Arrow arum (tuckahoe) Peltandra virginica Roots, stems, and leaves processed, boiled, and eaten Pickerelweed Pontederia cordata Roots, stems, and leaves processed, boiled, and eaten Broadleaf cattail Typha latifolia Root and tender young shoots eaten (also used medicinally) Greenbriar Smilax spp. Roots dried and ground for flour (also used medicinally) Food: Other Loblolly pine Pinus taeda Inner bark eaten as starvation food Eastern redbud Cercis canadensis Seed pods roasted and seeds eaten (also used for materials) Medicinal Allegheny chinquapin Castanea pumila Decoction of root used for fever (mast also used for food) Sassafras Sassafras albidum Root, root bark, and bark used in decoctions for many ailments Leaves and stems used in decoctions and teas for aches and Spicebush Lindera benzoin colds Bark chewed as toothache remedy, and mashed root used as Devil's walking stick Aralia spinosa poultice for skin ailments Materials White pine Pinus strobus Stems used for dugout canoes (also used ceremonially) Loblolly pine P. taeda Stems used for dugout canoes Bald cypress Taxodium distichum Bark used to make cordage Willow Salix spp. Bark used for weaving and basketry Grain storage; older trees with heart rot were known to be good Blackgum Nyssa sylvatica bee trees for honey

164 B. Literature utilized for ethnobotanical inventory

Abrams, M.D. and Nowacki, G.J. 2008. Native Americans as active and passive promoters of

mast and fruit trees in the eastern USA. The Holocene 18: 1123-1137.

Asch, D.C and Asch, N.B. 1977. Chenopod as cultigen: A re-evaluation of some prehistoric

collections from eastern North America. Midcontinental Journal of Archaeology 2: 3-45.

Arnason, T., Hebda, R.J., and Johns, T. 1981. Use of plants for food and medicine by Native

Peoples of eastern Canada. Canadian Journal of Botany 59: 2189-2325.

Briand, C.H. 2005. The common persimmon (Diospyros virginiana L.): The history of an

underutilized fruit tree (16th-19th centuries). Huntia 12: 71-89.

Day, G.M. 1953. The Indian as an ecological factor in the northeastern forest. Ecology 34: 329-

346.

Dweck, A.C. 1997. Ethnobotanical use of plants, Part 4: The American continent. Cosmetics and

Toiletries 112.

Pdf available online at http://www.dweckdata.com

Erichsen-Brown, C. 1989. Medicinal and Other Uses of North American Plants: A Historical

Survey with Special Reference to the Eastern Indian Tribes. Dover Publications, Inc.,

Mineola, New York.

Fernald, M.L. and Kinsey, A.C. 1943. Edible Wild Plants of Eastern North America. Idlewild

Press, Cornwall-on-Hudson, New York.

Furgerson, K.A. 2007. Archaeobotany of the Late Woodland and Contact Periods at the Barton

Site (18AG3), Allegany County, Maryland. Master of Arts, Archaeology and Heritage

thesis, University of Leicester, England.

Gremillion, K.J. 2004. Seed processing and the origins of food production in eastern North

America. American Antiquity 69: 215-233.

165 Hamel, P.B. and Chiltoskey, M.U. 1977. Cherokee Plants and Their Uses – A 400 Year History.

Herald Publishing Company, Sylva, North Carolina.

Hart, J.P., editor. 2008. Current Northeast Paleoethnobotany II. New York State Museum

Bulletin 572. The University of the State of New York Education Department, Albany,

New York.

Hart, J.P. and Rieth, C.B., editors. 2002. Northeast Subsistence-Settlement Change: A.D. 700-

1300. New York State Museum Bulletin 496. The University of the State of New York

Education Department, Albany, New York.

Havard, V. 1895. Food plants of the North American Indians. Bulletin of the Torrey Botanical

Club 22: 98-123.

Herrick, J.W. 1995. Iroquois Medical Botany. Syracuse University Press, Syracuse, New York.

Kuhn, R.D. and Funk, R.E. 2000. Boning up on the Mohawk: An overview of Mohawk faunal

assemblages and subsistence patterns. Archaeology of Eastern North America 28: 29-62.

MacDougall, A. 2003. Did Native Americans influence the northward migration of plants during

the Holocene? Journal of Biogeography 30: 633-647.

Messner, T.C. 2011 Acorns and Bitter Roots: Starch Grain Research in the Prehistoric Eastern

Woodlands. The University of Alabama Press, Tuscaloosa, Alabama.

Minnis, P.E., editor. 2010. People and Plants in Ancient Eastern North America. University of

Arizona Press, Tucson, Arizona.

Moerman, D. Updated summer 2003. Native American Ethnobotany: A database of plants used as

drugs, foods, dyes, fibers, and more, by native Peoples of North America. University of

Michigan-Dearborn.

http://herb.umd.umich.edu/

166 Parker, A.C. 1910. Iroquois Uses of Maize and Other Food Plants. New York State Museum

Bulletin 144. The University of the State of New York Education Department, Albany,

New York.

Peterson, R.N. 1991. Pawpaw (Asimina). In Genetic Resources of Temperate Fruit and Nut

Crops. Moore, J.N. and Ballington, J.R., editors. Acta Horticulturae 290: 567-600.

Scarry, C.M., editor. 1993. Foraging and Farming in the Eastern Woodlands. University Press of

Florida, Gainesville, Florida.

Tantaquidgeon, G. 2001. Folk Medicine of the Delaware and Related Algonkian Indians.

Anthropological Series Number 3, The Pennsylvania Historical and Museum

Commission, Harrisburgh, Pennsylvania.

Watson, P.J. 1989. Early plant cultivation in the Eastern Woodlands of North America. In

Foraging and Farming: The Evolution of Plant Exploitation. Harris, D.R. and Hillman,

G.C., editors. Unwin Hyman, London, UK. Pages 555-571.

VITA

Sarah E. Johnson

Education

The Pennsylvania State University, Ph.D., Ecology 2014 The Pennsylvania State University, M.S., Forestry 2007 Defiance College, B.S., Restoration Ecology 2005

Fellowships and Awards

McKenna Fellowship for Women in Science at Penn State Penn State College of Agricultural Sciences International Programs travel grant for South African Agroforestry Study Tour Funds for Excellence in Graduate Recruitment (FEGR) grant at Penn State McMaster School for the Humanities Research Fellowship at Defiance College for work in Belize, Central America

Publications

1. Brubaker, K., Johnson, S.E., Brinks, J., and Leites, L. Estimating canopy height of deciduous forests at a regional scale with leaf-off, low point density LiDAR. Canadian Journal of Remote Sensing, in review November 2013. 2. Abrams, M.D. and Johnson, S.E. The impacts of prescribed fire and mast year on tree regeneration in oak forests at the Mohonk Preserve, southeastern New York. Natural Areas Journal 33:427-434. 3. Abrams, M.D. and Johnson, S.E. Wildfire damage assessment in a young oak forest. Journal of Applied Fire Science, accepted October 2012. 4. Abrams, M.D. and Johnson, S.E. Long-term impacts of deer exclosures and land-use history on forest composition at the Valley Forge National Historical Park, Pennsylvania, USA. Journal of the Torrey Botanical Society, accepted January 2012. 5. Johnson, S.E. and Abrams, M.D. 2009. Age class, longevity, and growth rate relationships: protracted growth increases in old trees in the eastern United States. Tree Physiology 29: 1317- 1328. 6. Johnson, S.E. and Abrams, M.D. 2009. Basal area increment trends across age classes for two long-lived tree species in the eastern U.S. TRACE (Tree Rings in Archaeology, Climatology and Ecology) Proceedings 7: 127 - 134.