Prehistoric Colonization of Southcentral Alaska

Prehistoric Colonization of Southcentral Alaska:

Human Adaptations in a Post Glacial World

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in

Anthropology

Brian T. Wygal

Committee Advisor Dr. Gary A. Haynes

University of Nevada, Reno

May 2009

THE GRADUATE SCHOOL

We recommend that the dissertation prepared under our supervision by

BRIAN T. WYGAL

entitled

Prehistoric Colonization Of Southcentral Alaska: Human Adaptations In A Post Glacial World

be accepted in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Gary A. Haynes, Advisor

Michael R. Bever, Committee Member

Ted Goebel, Committee Member

David E. Rhode, Committee Member

G. Richard Scott, Committee Member

Scott A. Mensing, Graduate School Representative

Marsha H. Read, Ph. D., Associate Dean, Graduate School

May, 2009

Abstract

This dissertation presents results from four recently discovered archeological sites in southcentral Alaska. The sites range from the Younger Dryas to middle Holocene in age and provide valuable contextual information for the human process of colonizing a region that was heavily glaciated during the LGM. The deglaciation and human colonization of southcentral Alaska is one of the most significant aspects of the settling phase in eastern

Beringia not only for its potential to inform about the human response to post-glacial landscapes but also for what we can learn about subsequent migrations to the southern coast of Alaska.

Understanding how early foraging societies spread throughout eastern Beringia, after its initial colonization, requires new models for and approaches to the interpretation of technological variability, especially the enigmatic microblade industries that represent an essential subset of nearly all of the northern prehistoric toolkits. This work presents an ecological approach to the interpretation of assemblage variability in central and southcentral Alaska. Climate and seasonal changes had significant impacts on small- scale foraging societies and undoubtedly played a decisive role in the successes and failures of the earliest Alaskans. This was certainly the case during transition from the

Pleistocene to Holocene as major climatic oscillations were underway and foothold communities north of the Alaska Range began dispersing south into the recently deglaciated territories of southcentral Alaska.

Acknowledgements

This research was made possible by the National Science Foundation, Office of Polar Programs

Grant # 0520559. The grant proposal was prepared by Ted Goebel and me in 2004 and emerged from previous work by Fran Seager-Boss at the Matanuska-Susitna Borough, Cultural Resources

Division who was awarded a Federal Coastal Zone Management grant to conduct the 2004 survey. I am ever grateful to Fran Seager-Boss for placing me and an excellent crew (Kathryn

Krasinski, Dan Stone, and Randolph Tedor) in the position to discover the Trapper Creek sites and for continued support of further investigations by the University of Nevada, Reno (UNR).

Additional funding was provided by several UNR entities including the Sundance Fund, Graduate

Student Association, Summer Sessions Program, Academy for the Environment, and the

Department of Anthropology. The Center for the Study of the First Americans at Texas A&M and the Alaska Humanities Forum also provided funding for this research. Significant funding and support has come from the National Park Service, Alaska Regional Office, which provided a four-year Systemwide Archeological Inventory and Monitoring award that, in part, allowed a search for new sites related to my dissertation research in Denali National Park and Preserve.

Although still ongoing, the Denali project has substantially improved this research and for this I am extremely grateful. For making the Denali survey possible, I thank the former and current cultural resources managers Ann Kain and Amy Craver, archeologist Jeremy Karchut, and helicopter pilots George Houke and Shane Herron.

I sincerely thank all of the fieldschool students, volunteers, and team supervisors

(Kathryn Krasinski, Dan Stone, and Sam Coffman) who labored tirelessly and enthusiastically at these sites. The work of field and lab technicians Richie Bednarski, David Aonga, Jennifer

Kielhofer, Evan Pellegrini, Kelly Eldridge, and Travis Shinabarger is also greatly appreciated.

Although we may disagree on some aspects, I am grateful to Richard Reger, Randolph Tedor, and

Kelly Graf for their contributions and discussions on the sedimentology of the TCO site. iii

Acknowledgments are due for the specialized analyses by Elmira Wan at the USGS lab in Menlo

Park, Kristi Wallace with the Alaska Volcano Observatory, Anchorage, Steve Forman at the

Luminescence Dating Research Laboratory, University of Illinois, Chicago, and to Beta Analytic

and University of Arizona radiocarbon laboratories. Peter Wigand and his students have been conducting pollen analysis on the Trapper Creek bog cores and John McCormak, Department of

Geological Sciences and Engineering at UNR, provided assistance in confirming raw material

varieties and elemental characterization of grease residues.

I particularly want to thank Rodney “Norwood” Marsh and the community of Trapper

Creek, Alaska for their hospitality and for sharing their extensive local knowledge of the area.

Chad Valentine of Denali Anglers in Talkeetna provided river boat transportation to many survey areas throughout the Susitna River lowlands. Chad’s extensive knowledge of river currents and logistics were invaluable during the Trapper Creek phase of this research. Several specialists in

Alaskan archeology and geology have offered enlightening comments on my research including

David Yesner, Becky Saleeby, Charles Holmes, Richard VanderHoek, Bob Gal, Owen Mason,

William Workman, Douglas and Richard Reger, Don Dumond, and Richard Knecht. Although we may have differing interpretations of the data or hypotheses presented here, their discussions have provided valuable insights and greatly improved my research. I especially want to thank

Kathryn Krasinski for her patience and meticulous revisions on early drafts and also to my family for their unlimited support and encouragement.

I am thankful for my committee advisor, Gary Haynes, for his attention to detail and editorial skill. His efforts have greatly improved the overall organization and writing of this dissertation. I am also indebted to my graduate committee members Michael Bever, Ted Goebel,

Scott Mensing, David Rhode, and Richard Scott for their conversations and perceptive comments related to this research. Finally, I would like to thank my entire committee for their considerable support and advice which has been essential to my academic development. iv

TABLE OF CONTENTS

Abstract...... i Acknowledgments...... ii Table of Contents...... iv List of Tables...... vi List of Figures...... viii

I. INTRODUCTION……………………………………………………………….. 1 Environment and Setting………………………………………………….3 History of Research……………………………………………………….6 Central Alaska…………………………………………………………. 6 Southcentral Alaska…………………………………………………... 11 Conceptual Framework…………………………………………………. 17 Research Objectives…………………………………………………….. 20 Dissertation Organization……………………………………………….. 21

II. PALEOECOLOGY OF SOUTHCENTRAL ALASKA……………………….. 22 Deglaciation of Eastern Beringia……………………………………….. 24 Southcentral Alaska…………………………………………………... 27 The Aleutians and Southeast Alaska…………………………………. 28 Climatic Oscillations……………………………………………………. 29 Paleoecology of the Mammoth Steppe………………………………….. 32 Volcanism……………………………………………………………….. 39 Primary Evidence for Volcanism in the Trapper Creek Locality…….. 41 Conclusions……………………………………………………………... 44

III. ARCHEOLOGY OF THE SUSITNA RIVER LOWLANDS………………….. 46 Study Area………………………………………………………………. 48 Excavation History……………………………………………………… 49 Stratigraphy……………………………………………………...... 51 Methods………………………………………………………………. 52 Stratigraphic Analysis………………………………………………... 53 Dating…………………………………………………………………… 58 Artifact Assemblages…………………………………………………… 66 Methods………………………………………………………………. 67 Susitna River Overlook………………………………………………. 68 Trapper Creek Overlook, Component I……………………………….73 Trapper Creek Overlook, Component II……………………...... 77 Discussion………………………………………………………………. 80 Conclusions……………………………………………………...... 92

IV. UPLAND LITHIC PROCUREMENTAND HUNTING STRATEGIES……… 94 Study Area………………………………………………………………. 95 Research History………………………………………………...... 97 v

Stratigraphy and Dating…………………………………………………. 99 Bull River II…………………………………………………………... 99 Costello Creek………………………………………………...... 103 Artifact Assemblages…………………………………………...... 105 Methods……………………………………………………...... 105 Bull River II…………………………………………………………. 106 Costello Creek……………………………………………………….. 108 Discussion……………………………………………………………… 109 Upland Kill Sites……………………………………………………….. 116 Analogous Prehistoric Kill Sites…………………………………….. 117 Ice Patch Kill Sites in Alaska………………………………...... 118 Conclusions…………………………………………………………….. 122

V. THE MICROBLADE/NON-MICROBLADE DICHOTOMY: THE EFFECT OF CLIMATE AND SEASON ON TOOLKIT VARIABILITY…… 124 Microblade Technology………………………………………...... 126 Core and Blade Technologies in Eastern Beringia…………………….. 127 The Radiocarbon Record and Method of Analysis…………………….. 131 Methods of Evaluation………………………………………………. 132 Results……………………………………………………………….. 134 Prehistoric Populations and the Presence/Absence of Microblades...... 138 Microblades, Climate, and Ecology……………………………………. 141 Seasonality, Landuse, and Toolkit Variability…………………………. 143 The Gal Seasonality Model………………………………………….. 143 Annual Economic Model for Southcentral Alaska………………….. 147 Conclusions…………………………………………………………….. 150

VI. HUNTER-GATHERER TECHNOLOGICAL ORGANIZATION IN SOUTHCENTRAL ALASKA………………………………………………… 153 Intersite Statistical Analysis……………………………………………. 154 The Prehistoric Colonization of Southcentral Alaska: Human Adaptations in a Post-Glacial World…………………………………... 170 Implications for Major Theoretical Debates and Further Research in Eastern Beringia…………………………………...... 174 Conclusions…………………………………………………………….. 176

VII. APPENDIX…………………………………………………………...... 179

VIII. REFERENCES CITED………………………………………………………… 191

IX. CURRICULIUM VITAE.…………………………………………………….. 231

LIST OF TABLES

Table 2.1 Northern European pollen chronozones applicable in Alaska………………………. 24

Table 2.2 Summary of electron microprobe analysis, Trapper Creek bog 2005 core…………. 43

Table 3.1 Stratigraphy and dating of the Trapper Creek Overlook and Susitna River Overlook sites……………………………………………………………………….. 57

Table 3.2 Radiocarbon data from the Susitna River Overlook site…………………………….. 59

Table 3.3 Radiocarbon data from the Trapper Creek Overlook site…………………………….60

Table 3.4 Optically stimulated luminescence ages from TCO and SRO………………………. 63

Table 3.5. Raw material types at SRO………………………………………………………….. 69

Table 3.6. Raw material types at TCO-I………………………………………………………… 74

Table 3.7. Raw material types at TCO-II……………………………………………………….. 78

Table 3.8 Microblade raw material at TCO……………………………………………………. 87

Table 4.1 Radiocarbon data from the Bull River II site………………………………..……….103

Table 6.1 Summary of site settings, primary activities, lithic use, and potential season of occupation……………………………………………………………….. 155

Table 6.2 Intersite crosstabulation of flaked debris by raw material………………………….. 156

Table 6.3 Intersite crosstabulation of flaked debris by degree of dorsal cortex………………. 158

Table 6.4 Intersite crosstabulation of flaked debris by size class……………………………… 159

Table 6.5 Mann-Whitney tests on TCO-I and TCO-II degree of cortex and flake size among debitage……………………………………………………… 160

Table 6.6 Mann-Whitney tests on Bull River II and Costello Creek degree of cortex and flake size among debitage……………………………………………………… 160

Table 6.7 Flaked debris Tukey post hoc test of weight means…….………………………….. 163

Table 6.8 Tukey homogenous subsets of debitage weights…………………………………… 163

Table 6.9 Intersite crosstabulation of flaked debris by platform preparation……………..……165

Table 6.10 Intersite crosstabulation of debitage types……….………………………………... 167

vii

Table 7.1 Flaked debris frequencies at TCO and SRO………………………………………....179

Table 7.2 Tool class frequencies at TCO and SRO…………………………………………… 179

Table 7.3 Raw material type frequencies at TCO and SRO…………………………….…….. 180

Table 7.4 Artifact size class frequencies at TCO and SRO…………………………………… 180

Table 7.5 Raw material type frequencies at Bull River II and Costello Creek………………... 181

Table 7.6 Artifact size class frequencies at Bull River II and Costello Creek………………… 181

Table 7.7 Flaked debris frequencies at Bull River II and Costello Creek…………………….. 182

Table 7.8 Tool class frequencies at Bull River II and Costello Creek…………….…………... 182

Table 7.9 Evaluation of dated components in eastern Beringia and the presence or absence of microblades (methods of rejection)……………………………… 182

Table 7.9 Evaluation of dated components in eastern Beringia and the presence or absence of microblades (results)..……………………………………...…..... 183

viii

LIST OF FIGURES

Figure 1.1 Present-day Beringia with key features of southcentral and central Alaska…………..5

Figure 1.2 LGM ice extent and early period (>8 ka) archaeological components in Alaska……..6

Figure 2.1 Penultimate, LGM and modern glacial extent in southern Alaska…………………. 26

Figure 2.2 GISP2 Hulu delta 18O isotope climate curve after Weninger and Jöris (2008) and central Alaskan climate proxies……………………………………………….. 31

Figure 2.3 Known distribution of the Hayes set H and Oshetna tephra deposits………………. 41

Figure 2.4 Trapper Creek bog 2005 core…..……………………………………………….. 42

Figure 3.1 Susitna River lowlands with excavation histories of the TCO and SRO sites…….... 50

Figure 3.2 Stratigraphy and dating of TCO………………..…………………………………… 54

Figure 3.3 Stratigraphy and dating of SRO……..……………………………………………… 55

Figure 3.4 Distribution of point provenienced artifacts at SRO…..……………………………. 69

Figure 3.5 Refitting fragments of a biface from SRO……….…………………………………. 70

Figure 3.6 Heavy choppers from SRO including (a) a bifacial and (b) unifacial chopper with edge damage…….……………………………………………………………. 71

Figure 3.7 Point provenienced artifacts at TCO-I………..…………………………………….. 74

Figure 3.8 Lithic artifacts from TCO-I….……………………………………………………… 76

Figure 3.9 Point provenienced artifacts at TCO-II….…………………………………………. 78

Figure 3.10 Frequency of tool types (top) and flaked debris (bottom)………………………… 81

Figure 3.11. Refitting split cobble tci-thos implements from TCO-II……………….………… 82

Figure 3.12 Frequency of raw materials among tools at TCO and SRO……………………….. 85

Figure 3.13 Frequency of raw materials among flaked debris at TCO and SRO………………. 85

Figure 3.14 Degree of dorsal cortex on tools at TCO and SRO…………...…………………… 89

Figure 3.15 Degree of dorsal cortex on debitage at TCO and SRO…………………………….. 89

Figure 3.16 Tool size classes (cm) at TCO and SRO…….…………………………………….. 89

Figure 3.17 Debitage size classes (cm) at TCO and SRO..…………………………………….. 90

Figure 3.18 Tool platform preparation at TCO and SRO…….………………………………… 90

Figure 3.19 Platform preparation on flaked debris at TCO and SRO……..…………………… 91

Figure 4.1 Location of Bull River II and Costello Creek………………………………………. 96

Figure 4.2 Stratigraphic profile at Bull River II……………………………………….……….100

Figure 4.3 Carbon rich resinous stains at the base of Stratum 3 at Bull River II………………101

Figure 4.4 Elemental composition of organic stains from the Bull River II occupation horizon………..………………………………………… 102

Figure 4.5 Point provenienced artifacts and charcoal sample from Bull River II...……………103

Figure 4.6 Schematic stratigraphic comparison of Bull River II and Costello Creek.…………104

Figure 4.7 Stratigraphic profile and block diagram of point provenienced artifacts from Costello Creek……………………..…………………………………………….....105

Figure 4.8 Tci-thos and biface fragment from Bull River II…………….…………………….. 107

Figure 4.9 Frequency of tool types at Bull River and Costello Creek…...……………………. 110

Figure 4.10 Frequency of flaked debris at Bull River and Costello Creek…..………………... 110

Figure 4.11 Frequency of raw materials among tools at Bull River II and Costello Creek…… 112

Figure 4.12 Frequency of raw materials among flaked debris at Bull River II and Costello Creek……..………………………………………………………… 112

Figure 4.13 Degree of dorsal cortex on tools at Bull River II and Costello Creek……………. 113

Figure 4.14 Degree of dorsal cortex on debitage at Bull River II and Costello Creek…...…… 113

Figure 4.15 Tool size classes (cm) at Bull River II and Costello Creek…...………………….. 114

Figure 4.16 Debitage size classes (cm) at Bull River II and Costello Creek………..………… 114

Figure 4.17 Tool platform preparation at Bull River II and Costello Creek……….………….. 115

Figure 4.18 Platform preparation on flaked debris at Bull River and Costello Creek…..…….. 115

Figure 4.19 Projectile points and wood foreshaft from Basalt Lake Ice Patch 4……..………. 119

Figure 4.20 Ice patch above Bull River II and Costello Creek in Denali National Park……… 122

Figure 5.1 Microblade/non-microblade dichotomy and component frequency in eastern Beringia (1000-14,500 cal BP)…………………………………………… 137

Figure 5.2 Microblade/non-microblade dichotomy and component frequency in eastern Beringia (7700-14,500 cal BP)…………………………………………… 138

Figure 5.3 The annual economic model……..……………………………………………... 149

Figure 6.1 Frequency of raw materials among flaked debris from all components…………… 157

Figure 6.2 Frequency of debitage types among flaked debris from all components…………...168

Figure 6.3 Annual economic model for southcentral and central Alaska (12.5 to 6 ka)………. 172

CHAPTER I

INTRODUCTION

This dissertation is an investigation into the lives of the first human groups to enter and

use the various landscapes of southcentral Alaska. Alaska’s state motto, “the last

frontier”, is a particularly fitting slogan for today, but during the Pleistocene, Alaska was

actually the first frontier for the Paleolithic hunters who crossed into the Western

Hemisphere approximately 14.5 ka1 or earlier. Nevertheless, much of the state remained

unoccupied for millennia as it was not easily settled. This was particularly the case with

respect to its southern margins. As the descendents of Alaska’s first human groups

trickled south, vestiges of their technology and landuse strategies persisted into the

middle Holocene, but given the lack of reliably dated sites across the region, these are

events about which we know little.

Few places worldwide experienced shifts in paleoecology at the end of the last

glacial as radical as lands once covered by or adjacent to the massive ice sheets that

formerly blanketed much of the northern hemisphere. Climate clearly played a

significant role in determining the success or failure of the first settlements in Alaska.

Because foraging societies are, in fact, so much a part of their ecological settings (Kelly

1995), changes in their environment are the single most powerful force influencing

culture change (Harris 1968; Steward 1972; White 1959). This was probably the case in

Alaska where retreating glaciers and landscape recovery during the terminal Pleistocene,

1 When referring to general time (years before now) and not specific radiocarbon dates (years before 1950), calibrated radiocarbon ages have been rounded up to the nearest tenth and abbreviated (e.g. 20,740 cal BP is written 20.8 ka). This is done throughout the dissertation for consistency between disciplines and dating techniques.

17 to 10 ka, may have allowed foothold communities of people living north of the Alaska

Range to expand into previously unoccupied and unfamiliar landscapes south of the

Alaska Range—an event that was apparently interrupted by the onset of the Younger

Dryas cold snap 12.8 ka when significantly fewer dated occupations occur across the

state (Chapter V; Wygal 2007a, 2007b, 2008a, in press; Wygal and Krasinski 2006).

Understanding how these groups survived and spread after colonization requires some

understanding of debates over the typology and culture history of central Alaska before

addressing events in southcentral Alaska.

The first trace of human occupation of eastern Beringia dates between 14.4 and

13.6 ka at Swan Point where evidence suggests horse and mammoth were potential prey

species (Holmes in press). Initially, people were probably limited to interior central

Alaska where they endured the Older Dryas cold snap ca. 14 ka. A majority of the

earliest occupations date to the warmer Allerød period 14 to 12.8 ka at Dry Creek, Moose

Creek, Walker Road, and Owl Ridge in the Nenana valley (Goebel et al. 1991; Powers

and Hoffecker 1989; Ten Brink 1984) and Broken Mammoth Culture Zone (CZ) 3, Swan

Point CZ 3, and Mead at the Shaw Creek Flats in the Tanana valley (Holmes 2001, in

press; Péwé and Reger 1983; Yesner 1996, 2001).

Although we currently know most about the archeology of the Nenana and

Tanana valleys in central Alaska, answers may ultimately be found in the under-

investigated areas peripheral to these valleys. One such region is the southern slopes of

the Alaska Range and major river valleys of southcentral Alaska. Also of interest is the

human response to the changing post-glacial landscape of southcentral Alaska. How did

subarctic foragers spread throughout a vast territory previously free of human 3

inhabitants? What aspects of their traditional knowledge did they bring with them south

of the central Alaska Range? How were technology and settlement affected by extreme

environmental fluctuations during the terminal Pleistocene and middle Holocene?

Answers to these questions provide a broad view for how technology and economy

evolved and thus offer a greater understanding of prehistoric foragers than the traditional

culture-historic approach.

Situated between the Alaskan interior to the north and coastal areas to the south,

southcentral Alaska is well-suited for investigating the initial human expansion into

recently deglaciated landscapes. Unfortunately, the area remains relatively unknown

archeologically, particularly with respect to the area’s earliest occupants (Wygal and

Goebel 2006) so that answers to questions are often extrapolated from what we know of

the archeological record north of the Alaska Range. The first population expansion into

southcentral Alaska occurred during the late Pleistocene either from the north as glaciers

retreated further into the central Alaska Range as implied by Yesner (1998; 2001:316) and others (Ackerman 1992; Bacon et al. 1983; Dumond 1998:190; Moss 1998; Reger

1981, 1998), or from the south and west as the maritime environments of the Gulf Coast became deglaciated (Dixon 1993, 1999; Mandryk et al. 2001).

Environment and Setting

This dissertation focuses on the archeology of the Susitna River corridor from its upland headwaters near the west fork of the Chulitna River to the interior lowland region of southcentral Alaska. Southcentral Alaska is separated from central Alaska to the north by the arching central Alaska Range which has some of the highest peaks in North America.

Although the crest of the range is clearly defined, the drainage divide is more complex 4

with at least two major rivers (the Nenana and Delta) draining north through the

mountain chain via north-south trending mountain passes. For the sake of simplifying the

distinction between central and southcentral Alaska, the Alaska Range crest will define

this boundary rather than hydrological divide. Southcentral Alaska is more clearly defined on its southern boundary by the maritime environments of Cook Inlet and the

Gulf of Alaska.

Southcentral Alaska spans more than 40 million acres of varied environmental settings from near sea level to in excess of 6096 m (20,000 ft) in elevation. My research concerns two study areas along the George Parks Highway near the modern communities of Cantwell and Trapper Creek in the northwest corner of this expanse (Fig. 1.1). These two study areas represent entirely different topographic and biotic communities, providing contrasting ecological perspectives on the early prehistoric use of these landscapes. Each study area contains early period archeological sites that are the focus of this work. Specifically, the northern research area is in an alpine setting southwest of

Cantwell where the Bull River II and Costello Creek sites were named for the upper tributaries of the Chulitna River they overlook. To the south, the Trapper Creek

Overlook (TCO) and Susitna River Overlook (SRO) sites were discovered in the heavily forested middle Susitna River lowlands (Fig. 1.2).

Figure 1.1. Present-day Beringia with key features of southcentral and central Alaska. Red boxes along the George Parks Highway denote study areas and a dashed line delinates the conceptual northern boundary between southcentral and central Alaska. Map modified from AlaskaPak for ArcGIS 9.2 publically available from the National Park Service. 6

Figure 1.2. LGM ice extent and early period (>8 ka) archeological components in Alaska. Ice extent after Kaufman and Manley (2004) and subsequent human occupations in 2 ka intervals based on date evaluation outlined in Chapter V and Appendix Table 7.11. Sites coded according to earliest component. Sites and references are as follows: 1. Anangula (Laughlin 1975; Dumond and Knecht 2001), 2. Hog Island (Knecht and Davis 2001; Dumond and Knecht 2001), 3. Ugashik Narrows (Dumond 1975; Henn 1978), 4. Trapper Creek and Susitna River Overlooks (Wygal and Goebel 2006), 5. Ground Hog Bay II (Ackerman 1996a), 6. Hidden Falls (Davis 1989, 1996), 7. On Your Knees Cave (Dixon 1999), 8. Chuck Lake (Ackerman et al. 1985), 9. Bull River II (Wygal 2007c) and Carlo Creek (Bowers and Reuther 2008), 10. Tangle Lakes sites (West et al. 1996a, b, c), 11. Nenana valley sites (Powers and Hoffecker 1989), 12. Chugwater (Lively 1996), Broken Mammoth, Mead, and Swan Point (Holmes 2001; Yesner 1996), 14. Gerstle River (Potter 2005), 15. Batza Téna (Clark and Clark 1991), 16. Onion Portage (Anderson 1988), 17. Tululaq Hill (Rasic and Gal 2000; Rasic 2003), 18. Mesa and Lisburne (Kunz et al. 2003), 19. Tingmiakpuk (Mason et al. 2001), 20. Putu and Bedwell (Alexander 1987), 21. Spein Mountain (Ackerman 1996c), 22. Beluga Point (Reger 1981), 23. Crag Point (Jordan 1992). Map modified from AlaskaPak for ArcGIS 9.2 publically available from the National Park Service.

History of Research

Central Alaska. During the 1930’s, archeologists Nelson (1937) and Rainey (1940)

excavated the Campus site in Fairbanks and reported on a stone industry that was similar to Paleolithic technology in Northeast Asia based primarily on specific methods of 7

microblade production. Investigations continued in the Tanana valley where two

prehistoric components, an Athapaskan occupation and an earlier Campus-style core and

blade industry, were unearthed at Dixthada along Mansfield Lake (Lynch 1996:20;

Rainey 1940). In 1960, Forbes discovered the first two sites of archeological interest

north of the range divide near the Teklanika Campground in Denali National Park

(Morgan 1965). The discovery stirred excavations in 1961 and 1962 by West (1965) at

the Teklanika East and West sites.

The Campus style wedge-shaped cores discovered at the Teklanika sites were comparable to finds across central Alaska. Soon after, similar artifacts were discovered at Donnelly Ridge on the Ft. Greeley base in the Delta River valley. Similar tools were later discovered at Whitmore Ridge in the Tangle Lakes area located in southcentral

Alaska (West 1967, 1981; West et al. 1996c). West (1967, 1975, 1981, 1996b) included

Teklanika East and West as type sites in his “Denali complex”, what he believed to be the earliest evidence of people in Alaska. The Denali complex included microblades made from multifaceted wedge-shaped cores, specially prepared cores for large blade production, bifacial biconvex knives, flat-topped end scrapers constructed from thick flakes, burins, and burin spalls. Also diagnostic of the Denali complex were core tablets, a by-product of blade production, created when fresh platform surfaces were generated for blade removal (West 1967, 1975, 1981, 1996). Artifact comparisons with sites from

Siberia, especially the Dyuktai culture, led West (1975) to conclude the Alaskan sites dated between 12 and 10 ka.

In the 1990s, doubts were raised over the antiquity of West’s Denali complex finds (Dixon et al. 1985; Goebel 1992, 1996; Mobley 1991). In the work that ensued, 8

Mobley (1991, 1996) redated the Campus site to the mid-Holocene, and subsequent

radiocarbon dating on curated samples from Teklanika West suggested it could date as

late as 3.8 ka (West 1996b). Goebel (1992, 1996) revisited Teklanika West and identified three separate cultural horizons, but the earliest only dated to ca. 8 ka. The earliest secure dates on Denali complex assemblages occur at Dry Creek dated to 12.6 ka

(Hoffecker et al. 1996a). According to most recent accounts, during the early Holocene, the Denali industry continued seamlessly into what has typically been referred to as the late Denali complex (Dixon 1985; Mobley 1991), and it seemingly expanded geographically into northern and southern Alaska (Ackerman 1992; Mason et al. 2001).

Following the original Teklanika excavations, surveys along the Teklanika and

Nenana River valleys yielded many historic and prehistoric sites including a stratified late

Pleistocene site, Owl Ridge (Plaskett 1976; Lynch 1996; Morgan 1965, Treganza 1964).

North of Healy along the Nenana River, the most comprehensive archeological survey in central Alaska was inspired by earlier research in the area (Powers and Guthrie 1977).

The North Alaska Range Project led to a series of supplemental programs aimed at understanding late Pleistocene archeology and geology in the Nenana valley and the

Front Range, the northern foothills of the central Alaska Range (Goebel et al. 1991;

Phippen 1988; Pearson 1999; Powers and Hoffecker 1989; Ten Brink and Waythomas

1985). Sites found in the vicinity of Healy in the Nenana valley led to the definition of

the Nenana complex as the earliest Alaskan lithic industry (Goebel et al. 1991; Powers

and Hoffecker 1989).

The Nenana complex was dated between ca.13.2 to 13 ka (Powers and Hoffecker

1989) through correlated stratigraphy and a suite of 14C samples from the lowest cultural 9

components of Dry Creek (Powers et al. 1983; Powers and Hoffecker 1989; Powers and

Hamilton 1978), Moose Creek (Hoffecker 1996; Hoffecker and Powers 1985; Pearson

1999), Walker Road (Goebel et al. 1991; Goebel et al. 1996), and Owl Ridge (Phippen

1988). Diagnostic attributes of the Nenana complex include core and blade primary mode of reduction and characteristic tear-drop to triangular-shaped bifacial projectile points or knives in addition to large retouched blades (Goebel et al. 1991). The points were similar to those Cook (1969, 1996) termed Chindadn from Healy Lake in the

Tanana valley. The points frequently appear as expediently produced tools with a limited amount of bifacial trimming. Other, larger Chindadn style points are less common and considerably well made; these tools probably functioned as knives. Recent finds at the

Little John site in the Alaska/Yukon borderlands also include Chindadn-like bifaces

(Easton et al. 2008). For some archeologists (Carlson 1991; Goebel et al. 1991; Goebel

2004; Haynes 2002; Yesner 1996), the Nenana complex was considered a likely ancestor for Clovis, the earliest well-dated Paleoindian complex in lower North America.

The Nenana complex is chronometrically earlier and occurs in stratigraphically lower layers than Denali components in the Nenana valley (Powers and Hoffecker 1989).

Based on clear evidence for stratigraphic separation between the occupations, they were reasonably assumed to be separate technological traditions (Bigelow and Powers 2001;

Powers and Hoffecker 1989:272; cf. Thorson 2006). Consistency between several multi- component sites indicated the Nenana complex lacked the distinctive microblade production associated with an apparent second wave of colonizers represented by West’s

(1967, 1975, 1981) Denali complex. 10

In the Tanana valley, Healy Lake produced Chindadn points reportedly in

association with microblade technology (Cook 1969, 1975, 1996); however, the site was

excavated in arbitrary levels and correlating radiocarbon samples between units was not

always possible. In 1990, the Shaw Creek Flats, Tanana valley, received considerable

attention after a proboscidean tusk point was found at Mead (Péwé and Reger 1983).

Soon, similar late Pleistocene discoveries were made at Broken Mammoth (Yesner 1996;

2001) and Swan Point (Holmes in press; Holmes et al. 1996). These sites contained

stratigraphically sealed components now considered the earliest in the state.

Among the discoveries in the Shaw Creek Flats sites was the dating of “Old

World-style” microblade technology to the end of the Bølling interstadial (14.4 and 13.9

ka) at Swan Point CZ 4 (Holmes 2001, in press; Holmes and Crass 2003). This re-

opened the debate regarding the culture history of the earliest people in Alaska. Artifacts

uncovered in the late Pleistocene CZ 4 and early Holocene CZ 3 occupations at Broken

Mammoth were dominated by well-preserved fauna, but the small lithic assemblage from

CZ 4 was not diagnostic of any of the traditional complexes defined in the Nenana valley

despite their comparable ages (Krasinski 2005; Yesner 1996, 2000, 2001; Yesner et al.

1992; Yesner et al. 2000). Except for Swan Point CZ 4, the earliest evidence of

microblade production in the Tanana valley dates to ca. 12.1 to 11.9 ka, where it appears alongside small triangular Chindadn-style points from Broken Mammoth CZ 3 (Holmes

1996, 2001, in press; Krasinski 2005; Yesner et al. 1992), Swan Point CZ 3 (Holmes

2001, in press), and probably in the lower levels of Healy Lake (Cook 1969, 1975, 1996).

Although these occupations are coeval with Dry Creek component II, the “triangular to sub-triangular points” (Krasinski 2005:31) are reminiscent of the older Nenana complex. 11

If the earliest humans in eastern Beringia brought microblade technology from

Siberia, how do we explain the microblade/non-microblade dichotomy in eastern

Beringia (Hoffecker 2001:150)? Both Holmes (2001, in press) and West (1996a:550) reject the Nenana complex as defined by Powers and Hoffecker (1989) and Goebel et al.

(1991), preferring to include the Nenana complex within the Eastern Beringian tradition, but that approach still does not explain the lack of microblade technologies within the lowest levels of the Nenana valley sites. Suffice it to say, the Nenana complex seems to represent the initial dispersal of humans into today’s “Parks Highway corridor” of the

Nenana River valley, and from this time the northern foothills of the Alaska Range were regularly occupied by hunter-gatherers.

Southcentral Alaska. Unanswered questions remain about when and in what form humans began using the diverse landscapes south of the Alaska Range divide. Important alpine and montane sites including Phipps, Whitmore Ridge, Sparks Point, Carlo Creek,

Eroadaway, and Jay Creek Ridge (Fig. 1.2) have been documented (Bowers and Reuther

2008; Dixon 1993, 1999; Goebel 1996; Holmes 1988; West et al. 1996a, b, c). Prior to this research, however, no early period (>8 ka) sites had been reported in the heavily forested river lowlands in southcentral Alaska and there are precious few potentially early sites from coastal regions: Beluga Point along Cook Inlet, Ugashik Narrows on the

Alaska Peninsula, and Crag Point on Kodiak Island (Dumond 1975, 1998; Henn 1978;

Jordan 1992; Reger 1981, 1996, 1998).

The Tangle Lakes area is unique because although the setting is technically on the south slope of the Alaska Range, it is also in close proximity to the upper Delta River basin which drains to the north into the Tanana River. However, because many other 12

creeks in this region drain into the Susitna River, the sites from this area are considered a

part of southcentral Alaska. Phipps, Whitmore Ridge, and Sparks Point are three

shallowly buried sites located in the alpine Tangle Lakes area. In all three sites, lithic artifacts that West et al. (1996a, 1996b, 1996c) assigned to the Denali complex were found in sediments containing small amounts of dispersed charcoal dated between 12.5 and 9 ka. The shallow and potentially disturbed contexts of the samples make the dating of these sites equivocal. West (1996a:549) himself questioned the veracity of the ages for dating the associated cultural remains, concluding, largely on typological grounds, that the artifacts pre-dated 10.8 ka. Given that wedge-shaped core and microblade

technologies have been shown to persist through the middle Holocene in central Alaska

(Mason et al. 2001; Mobley 1991; Wygal in press), these Tangle Lakes sites are just as

likely to post-date 10 ka. Minimally, these three occupations were deposited prior to the

development of overlying paleosols dated to approximately 8 and 5.7 ka, respectively.

Clearly, the lack of precise proveniences, contextual data, and site formation analyses in

these shallow alpine sites makes it difficult to unequivocally date the artifacts.

Bowers (1978, 1979, 1980) unearthed two early cultural components (I and II) in

the deeply buried (ca. 3.5 m) Carlo Creek site located south of Nenana River gorge and

20 km north of Cantwell. Because of its location, Carlo Creek is considered to be just

inside the geographic boundary for southcentral Alaska. In the 1970s, one hearth

produced ages of 8400 ± 200 and 10,040 ± 440 BP, and the other an age of 8690 ± 330

BP (Bowers 1980). Bowers (1980:98) concluded that component I dated to ca. 9.6 ka

(8500 BP). Recently, Bowers and Reuther (2008) used the more sensitive AMS

technique to date curated samples from the hearths and a caribou bone recovered from 13

component I. Together, these suggested an age of 11.35 ka (11,300 ± 50 cal BP / 9966 ±

55 BP)2 for the earliest occupation at Carlo Creek.

Artifacts associated with the Carlo Creek hearths include a small set of complete and fragmented bifaces described as preforms or choppers, a large prismatic blade or

blade-like flake, cobble anvilstone, several thousand flakes, and three possible bone tools.

Although these latter artifacts were poorly preserved, two of the bone implements were

interpreted as distal point fragments made from a caribou metapodial. Lithics associated

with a hearth bore signs of heat treatment, leading the investigators to interpret part of the

lower component as a specialized lithic activity area. No microblades were found but the

assemblage was attributed to the Denali complex based on similarities to a special

activity area at Dry Creek component II (Bowers and Mason 1992:39). Based on its

stratified nature, dated cultural features, and location within the central Alaska Range,

Carlo Creek became perhaps the most significant site connecting central and southcentral

Alaska.

Another site located in the upper Nenana valley, Eroadaway, was found in the late

1980s and contained two shallowly buried cultural features, presumably hearths, and discrete concentrations of lithic artifacts. A bulk charcoal sample from one feature dated to 8640 ± 170 BP suggested the occupation occurred sometime between 10 and 9.5 ka.

The lithic assemblage is characterized by small, thin bifacial points and knives and associated bifacial reduction debris (Holmes 1988). Faunal remains included fragments

2 Radiocarbon years (BP) were calibrated into calendar years before present (cal BP) at 1σ using the CalPal (Weninger 1986; Weninger et al. 2005) calibration software, May 2006 version (CalPal 2006), and the Intcal04 curve (Reimer et al. 2004). 14

of burned medium-to large-bodied mammal bone. Holmes (1988) interpreted the short- term occupation as a preliminary game dismemberment and toolkit refurbishing site.

In the early 1980s, the Susitna River hydroelectric archeological survey project resulted in the discovery and excavation of Jay Creek Ridge (Dixon 1999; Dixon et al.

1985; Dixon and Smith 1990). Jay Creek Ridge was dated from dispersed charcoal samples, suggesting an age of about 8.1 ka. Stratigraphically, the cultural component occurred below a tephra layer attributed to the Oshetna eruption independently dated to ca. 6750-5850 cal BP (Dixon 1993). The site was tentatively assigned to a late phase of the Nenana complex based on the presence of concave-based bifacial points and a lack of microblades (Dixon 1993, 1999; Dixon et al. 1985).

Besides this small set of 14C-dated sites in southcentral Alaska, there are also numerous surface and near-surface sites that have not been 14C dated but contain typologically unique artifacts (Mobley and Morris 1981; Reger and Bacon 1996;

Robinson et al. 1996; West 1972, 1973, 1981, 1996b). One of these is the Amphitheater

Mountain complex, which West (1974) proposed to classify a group of similar surface scatters in the upper Delta River valley near Tangle Lakes. West (1975) dated the

Amphitheater Mountain complex to more than 12 ka (1975:80) based on the

“Mousteroid” nature of the tools. Included in the assemblage were large “pear shaped or foliated” bifaces and large “blade-like” tools recovered in situ in tightly clustered arrangements. The artifacts were “archaic in appearance” and heavily weathered, and diagnostic artifacts from other industries (i.e. Denali or Northern Archaic) did not occur within the Amphitheater Mountain complex sites (West 1974:219). Although this designation has been disputed (Bever 2001b:139; Mobley 1982; Yesner and Pearson 15

2002:139), nearby assemblages within the Alaska Range are comparable, including

materials thought to be deposited in butchering areas reported from component II at Dry

Creek (Powers et al 1983) and Carlo Creek (Bowers 1979).

Along Turnagain Arm south of Anchorage, Reger (1981, 1996, 1998) recovered

microblades, microblade core fragments, and a biface fragment from the lowest

component of Beluga Point, a stratified, multi-component site. Although not directly

dated, Reger (1996) suggested an early Holocene age based on similarities with the

Denali complex. Typologically, the assemblage is most reminiscent of the Ugashik

Narrows assemblage (Henn 1978; Reger 1981:181).

Besides Beluga Point, to find potentially early Holocene evidence of humans in

the maritime regions of southern Alaska, one has to look beyond the Susitna basin and

consider evidence from the Alaska Peninsula and Kodiak Island. The Ugashik Narrows

site (≤ 10 ka) is a stratified, multi-component site with an early assemblage most

reminiscent of the Akmak component at Onion Portage with similarities to the Denali

complex of interior Alaska. The Ugashik site is the most southerly of those classifiable

under the broad umbrella of the Paleoarctic tradition. Based on its geographic position, it

is a good ancestral candidate for the subsequent emergence of the Anangula tradition in

the Aleutians (D. Dumond 2009 pers. comm.).

The Ugashik assemblage contained microblades, large blades, and some bifacial

artifacts, and it yielded three conventional 14C ages, 7675 ± 260, 8425 ± 115, and 8995 ±

295 BP (8560 ± 320, 9390 ± 120, 10,120 ± 390 cal BP) in association with artifacts and a possible structure (Dumond 1975, 1998; Henn 1978; Reger 1981). The youngest of these dates (7675 ± 260) is suspect because it derived from dispersed charcoal. More reliable 16

are the two remaining ages suggesting the Ugashik occupation could predate Anangula

by as much as 500 years. Located away from the coast along a probable caribou migration route, Ugashik has dates that were probably not affected by problems associated with the use of driftwood (D. Dumond 2009 pers. comm.), which has been shown in some late Holocene sites on the Aleutians to predate human occupations

(R. Knecht 2009 pers. comm.). Another issue with comparing the dates of coastal sites to those in the interior involves the marine carbon reservoir effect which can produce age discrepancies as high as 400 14C years from terrestrial carbon sources (Stuiver et al.

1986).

Crag Point is an extensive, stratified, multi-component site located on Kodiak

Island, buried beneath a series of stratified Kachemak tradition occupations. Jordan

(1992) dated charcoal collected from a potential hearth feature in a basal cultural

component to 8.8 ka (8750 ± 680 cal BP / 7790 ± 620 BP) and attributed the assemblage

to the late “Paleoarctic tradition”, i.e., Denali complex (Jordan 1992). Others have

doubted the accuracy of the early Crag Point date because it came from a small sample

with a relatively large margin of error (Mills 1994). Doubts were raised over the nature of the lower assemblage as well, since it was associated with artifacts more characteristic

of the later Ocean Bay tradition (Steffian et al. 2002:4). The remaining maritime sites in

southcentral Alaska suggest colonization of the coast occurred 1000 years later, ca. 7.6 to

7 ka (Steffian et al. 2002).

The inland-migration-source hypothesis for the initial peopling of southcentral

Alaska is analogous to inquiries concerning the earliest inhabitants of southeast Alaska

where similarities between the Northwest Microblade tradition and Denali complex are 17

apparent. Microblades occurred in the basal components at Ground Hog Bay 2, Hidden

Falls, On Your Knees Cave, and Chuck Lake sites where the earliest dates reach 10.3 ka.

Most specialists in the area agree there is a clear transition from an interior tool kit and subsistence base to the later maritime cultures of the Northwest Coast (Ackerman 1992,

1996b; Dixon 1999; Moss 1998).

The sites discussed above provide a general, albeit limited, perspective on early human use of the various environments of southcentral Alaska during the early Holocene.

All of the alpine sites are situated at fairly high elevations (>853.4 m / >2800 ft asl) and have been interpreted to represent short-term seasonal occupations because these areas would have been largely inhospitable during winter months (Wygal 2008a, 2008b).

Eventually, foragers shifted their economy more toward the broad riverine settings of the

Susitna valley, and by the late Holocene annual salmon runs emerged as the primary subsistence strategy (Potter 2008a).

Conceptual Framework

Human expansion and colonization occurred at different rates during the Pleistocene-

Holocene transition, and questions surrounding when and how early pioneers first entered the unfamiliar landscapes of Beringia and ultimately the Americas remain central in

Arctic archeology today (Akazawa 1996; Anderson and Gillam 2000; Barton et al. 2004;

Bonnichsen and Turnmire 1999; Brantingham et al. 2004; Dixon 1999, 2001; Dumond

2001; Goebel 1999, 2002, 2004; Goebel et al. 1991, Goebel et al. 2003; Goebel and Buvit in press; Graf 2008; Greenberg et al. 1986; Haynes 2002; Hoffecker 2002; Hoffecker et al. 1993; Kelly 2003; Klein 2000; Kunz et al. 2003; Kunz and Reanier 1994, 1995;

Kuzmin and Tankersley 1996; Kuzmin and Orlova 1998; Larichev et al. 1988; 18

MacDonald 2004; Madsen 2004; Mandryk et al. 2001; Matson and Coupland 1995;

Meltzer 2002, 2003, 2004; Powers 1973; Roosevelt 2000; Schurr 2004; Schurr and

Sherry 2004; Vasil'ev et al. 2002; West 1996a; Wilson and Burns 1999; Yesner et al.

2004; Yi and Clark 1985).

Theoretical dynamics involving similar processes in the colonization of unfamiliar landscapes have been considered. Specifically, Beaton (1991), Kelly (1992), and Kelly and Todd (1988) examined conditions of mobility and the colonization of new territories. Meltzer (2002, 2003, 2004) has speculated on resource learning in new landscapes while MacDonald (1998) modeled population size requirements for maintaining successful settlements in new territories. Yesner (1996, 1998) proposed a

“push-pull” model where environmental degradation and submergence of the Bering

Land Bridge pushed populations out of central Beringia (Hoffecker and Elias 2007), and an abundance of mammal and avian species in the Tanana valley enticed foragers to migrate even further east. Mason et al. (2001) speculated the onset of cold climates in interior Alaska fostered an expansion of early Holocene colonizers, but others suggest world-wide cold oscillations caused population declines among Beringian human populations (Graf 2005, 2008; Wygal 2007a, 2007b, 2008a, in press).

During turbulent late glacial conditions, eastern Beringia suffered a declining steppe-tundra environment, and by 13 ka much of the ice mass that previously blanketed southern Alaska was limited to mountain valleys (Briner and Kaufman 2008; Dortch

2006). Shifts in tundra vegetation may have forced late Pleistocene megafauna species into enclaves of familiar refugia within a 30 km stretch of foothills bordering the northcentral Alaska Range (Ager 1975:85-86). Paleolithic hunters adjusting to declining 19

game may have converged on these remaining megafaunal populations (Ten Brink and

Waythomas 1985; Powers and Hoffecker 1989:263, 271). Soon after, as the mountain

passes of the Alaska Range opened and uplands became deglaciated, humans dispersed

further into these areas in search of prey such as caribou and Dall sheep. Based on what

is known thus far, Alaska’s southern coasts remained unpopulated until ca. 10 ka

(Ackerman 1992; 1996a; 1996b; Ackerman et al. 1979; Dumond and Bland 1995;

Dumond and Knecht 2001; Knecht and Davis 2001; Laughlin 1975; Moss 1998; Reger

1981, 1996, 1998; Veltre et al. 1984; Yesner 1998, 2001).

North of the Alaska Range, many early assemblages contain microblades and microblade byproducts among other tools, while others do not. Different approaches have recently been used to attempt to resolve this issue of assemblage variability in eastern Beringia (Bever 2006; Esdale 2009; Goebel and Buvit in press; Odess and Rasic

2007; Rasic 2008). What can the archeological record south of the Alaska Range tell us about this distinction? Is there a detectable pattern behind why humans discarded microblades at some places but not others? Early period foragers in interior Alaska had ancestry from the Siberian mammoth steppe and were highly mobile and accomplished hunters (Graf 2008) who probably utilized different elevation zones to extract seasonally available resources. Were the various environments south of the Alaska Range used in this way? Can seasonality be gleaned from the archeological record?

Essential to interpreting the early prehistory of southcentral Alaska are the geological and paleoecological transitions since deglaciation. How long did the deglaciation process take before the landscape was suitable for large mammals, including people? Was the broad Susitna valley teaming with bison, wapiti, caribou, sheep, and 20

potentially mammoth following the last glaciation? Did deteriorating climates in the interior force groups south toward the coast as major river valleys emerged from a once formidable ice sheet (Yesner 2001), or did southern Alaska’s first human residents originate from coastal groups already in place during the late Pleistocene (Dixon 1999)?

Important for understanding post-glacial human adaptations in southcentral

Alaska are questions related to their technological organization and landuse. How did the colonization process unfold? Was there an exploratory phase? Was there a lengthy period where southcentral Alaska was a marginal backwater for early foraging societies?

Did early foragers occupy the river lowlands of southcentral Alaska on a full-time basis or did small logistical hunting parties utilize the region seasonally? Finally, can sense be made of the enigmatic Amphitheater Mountain complex? Does it really represent an early Paleolithic occupation in the mountains of southcentral Alaska, or are the archaic- looking tools a facies of Holocene industries, perhaps a non-microblade subset of the

Denali complex? Investigations of archeological sites in southcentral Alaska allow for the critical evaluation of these unanswered questions which can be addressed through a systematic analysis of prehistoric technology, landuse, and chronology in central and southcentral Alaska.

Research Objectives

The primary research objective has been to locate and investigate previously unknown archeological sites that add to our understanding of the earliest migrations into the river lowlands of southcentral Alaska. Lithic assemblages from these sites can be used to determine how early foragers organized their technology south of the Alaska Range and why microblades appear in some sites but not others. In conjunction with proxy climate 21

records, an important objective is examining what became of these groups as central and southcentral Alaska faced increasingly warm temperatures and the mass extinction of primary food sources. Instead of focusing on placing new assemblages into old cultural historical models, I have compared climatic proxy records to the radiocarbon chronology and developed a seasonality model in an attempt to seek environmental and landuse influences on assemblage variability and occupation intensity in eastern Beringia.

Dissertation Organization

This dissertation is arranged in six chapters. Following this chapter is a review of the glacial chronology and paleoenvironments of southcentral Alaska during the late glacial and early post-glacial times (Chapter II). Chapters III and IV present the results of field investigations of the Trapper Creek Overlook and Susitna River Overlook sites located in the middle Susitna River lowlands as well as the Bull River II and Costello Creek sites found in the uplands southwest of Cantwell. These chapters include a description of detailed lithic, stratigraphic, and other analyses from these sites. Chapter V considers these sites in the wider context of the early prehistory and paleoenvironments in eastern

Beringia. Chapter VI uses statistical approaches to empirically demonstrate how the lithic data from excavated deposits relate to hypothetical models outlined in previous chapters. Finally, because so much of Alaskan archeology is concerned with reconciling the presence or absence of microblade technology in early sites (Bever 2006), I end with a new and testable theoretical perspective on the microblade-non/microblade debate based on the seasonal use of upland and lowland settings in central and southcentral

Alaska. 22

CHAPTER II

PALEOECOLOGY OF SOUTHCENTRAL ALASKA

The now submerged Bering Land Bridge was the ancient link between Alaska and eastern Siberia; it formed during periods of low oceanic levels caused by Pleistocene glaciations (Hopkins 1959, 1967). Greater Beringia is commonly defined as the region between Siberia’s Lena River and the Mackenzie River in the Northwest Territories of

Canada (Guthrie 2001; Hoffecker and Elias 2007). During the Last Glacial Maximum

(LGM) ca. 22 to 19 ka, eastern Beringia was less glaciated than most of northern North

America, with the majority of ice covering southern Alaska and the Brooks Range

(Briner and Kaufman 2008; Coulter et al. 1965; Hamilton and Thorson 1983; Kaufman and Manley 2004; Schmoll and Yehle 1986).

The increase in glacial ice hoarded much of the Earth’s freshwater during the late

Pleistocene causing an increase in atmospheric aridity, a lowering of worldwide sea levels, and the exposure of the continental shelf that is now beneath the Bering and

Chukchi seas. At the end of the Pleistocene, rapid melting of massive ice sheets throughout the Northern Hemisphere released large volumes of fresh water into the world's oceans causing sea levels to rise by as much as 120 meters. As a result, central

Beringia was flooded between 12 and 11 ka, severing the link between the Old and New

Worlds (Hopkins 1967).

In this chapter, I examine the changes in paleoenvironmental conditions from the

LGM to the middle Holocene, especially in central and southern Alaska.

The prehistoric environments and glacial chronologies of Alaska, beginning with the Penultimate stadial of the Wisconsin glaciation, and continuing through the LGM to

the middle Holocene, are significant because the process of deglaciation and

paleoecological responses which followed are integral to understanding the dispersal of

early humans in eastern Beringia. The timing of glacial cycles, paleoecology, climate,

and volcanism provides the context for the initial human colonization and subsequent

adaptations in eastern Beringia, particularly in southern and central Alaska. Although

microclimates led to environmental and ecological variations between regions, the most

substantial climatic changes in southcentral Alaska coincide with worldwide trends first

documented in pollen chronozones in northern Europe (Mangerud et al. 1974; Mangerud

and Berlund 1978), and are supported by proxy records for annual average temperatures

reconstructed from Greenland ice cores (Walker et al. 1999).

A summary of this chronology appears in Table 2.1 along with event horizons

documented in the Greenland Ice Core Project (GRIP and GISP2). The widely reported

discrepancy between terrestrial radiocarbon years and those from the ice layers are the

result of radiocarbon variability. At times there were sharp increases and decreases in the

density of atmospheric radiocarbon (14C) during the late glacial thereby causing slight

discrepancies in age determinations from different times (Alley et al. 1993; Fiedel 1999,

2000). During the Pleistocene-Holocene transition, 14C dates tend to be ca. 2000 years

younger than calendar years before present. For clarity, both the calibrated and

uncalibrated radiocarbon dates from terrestrial sites and dates taken directly from GRIP

ice horizons are presented here, along with corresponding chronozones and a description

of typical vegetation cover from southern Fennoscandia, an extensively researched region that underwent post-glacial processes analogous to southern Alaska (Wygal in review).

Because some of the European terminology has been adopted by Alaskan researchers, I

apply the system, where applicable, to the ecological transitions of southcentral Alaska; however, there is some variation between timing of chronozones in Alaska from those in

Northern Europe.

Table 2.1. Northern European pollen chronozones applicable in Alaska. Chronozone 14C BP cal BP GRIP BP GRIP Event Vegetation Interstadial 19.1-18 22.6-21.3 21.8-21.2 GI 2 Sparse steppe tundra LGM 18-16 21.3-19.2 21.2-19.5 GS 2c none/sparse Late glacial 16-14 19.2-16.7 19.5-16.9 GS 2b Sparse/steppe/tundra Late glacial 14-13.3 16.7-15.8 16.9 -14.7 GS 2a Steppe/herb/tundra Bølling 13-12 15.35-13.9 14.7-14 GI 1e Steppe/herb/tundra Older Dryas 12-11.8 13.9-13.7 14-13.9 GI 1d Steppe/shrub/tundra Allerød 11.8-11 13.7-12.95 13.9-12.9 GI 1c-1a Parkland, sparse birch forest Younger Dryas 11-10 12.95-11.5 12.7-11.5 GS 1 Parkland/tundra Preboreal 10-9 11.5-10.1 11.5-10.1 ― Open birch, pine, aspen forest Boreal 9-8 10.1-8.9 10.1-8.8 ― Mixed birch, white spruce forest From Björck et al. 1998; Eriksen 2002; Fiedel 1999; Gamble et al. 2004; Haynes 2002; Mangerud et al. 1974; Walker et al. 1999. Dates are abbreviated thousands of years. Calibrations by CalPal (Weninger et al. 2005) software at one sigma with the intcal04 curve (Reimer et al. 2004).

Following the LGM, dramatic shifts in annual average temperatures and climate

caused a series of glacial readvances and retreats. Although these were more extensive

than modern or recent ice flows, they were primarily limited to mountain valleys and

foothills, and were more restricted than the piedmont ice domes of the LGM. Moreover,

landscapes were altered significantly by shifts in floral and faunal communities and

increased volcanic activity in the post-glacial period, the latter depositing a series of

tephra layers across southern and central Alaska.

Deglaciation of Eastern Beringia

During full-glacial times of the late Pleistocene, the Talkeetna-Trapper Creek area was

mantled by ice flowing from high mountain valleys of the Talkeetna and Alaska ranges

(Hamilton and Thorson 1983:38), with three major ice flows coalescing from directions

mirroring the modern courses of the Chulitna, Susitna, and Talkeetna rivers. The precise

timing of deglaciation in the middle Susitna River lowlands is not well understood

(Schmoll and Yehle 1986), making it difficult to assess the human response to ice retreat following the LGM. As a consequence, models based on evidence from neighboring

regions are required for understanding late glacial and ecological chronologies of the middle Susitna and Chulitna River study areas. Simulation models have indicated that

maximum ice sheet volume in southern Alaska may have been reduced by 15% at 16 ka

(ca. 13,500 BP) and by 60% at 14 ka (ca. 12,000 BP) (Kutzbach et al. 1998:475).

Research in the northern valleys of the Alaska Range, particularly in Denali National

Park and the adjacent Nenana valley, has been extensive and offers physical traces for the timing of late Pleistocene glacial cycles in southern Alaska (Dortch 2006; Ten Brink

1984; Ten Brink and Waythomas 1985; Werner 1982; Werner et al. 1993).

Remnant moraine features preserved in the river valleys of the Front Range document the timing and extent of fluctuating glaciers. Of these, the Muldrow glacier near Wonder Lake in Denali National Park has been well-studied (Briner and Kaufman

2008; Harrison 1964; Kaufman and Manley 2004; Ten Brink and Waythomas 1985;

Viereck 1966; Weertman 1962). As the Muldrow glacier waxed and waned in the

McKinley valley, it left moraine features that were used as benchmark indicators for the local McKinley Park (MP) late glacial stadials. Because these correlate with moraine features in the Swift Fork (Kuskokwim), Heron, Foraker, Nenana, and Delta River valleys, the MP cycles are considered representative of glacial cycles in the central

Alaska Range.

Glacial advances during the LGM (equivalent to the MP I stadial) were not as

extensive as those of the Penultimate (60-55 ka) in central Alaska (Briner and Kaufman

2008; Hamilton and Thorson 1983; Ten Brink and Waythomas 1985; Wahrhaftig 1958).

During the LGM, the MP I glaciers advanced north from the Front Range as far as the

present location of the Denali Park road where terminal moraines were deposited between

21.4 and 20.6 ka. The MP II advance occurred 20.6 and 19.9 ka and was nearly as

extensive as MP I with ice covering southern portions of Wonder Lake but falling short

of the park road (Fig. 2.1).

Figure 2.1. Penultimate, LGM and modern glacial extent in southern Alaska. Depicted are the glacial limits during the “Healy” penultimate glaciation dated to 60 and 55 ka (blue line), LGM, 22 to 19 ka (shaded in pink), and modern glaciation (white). Key river valleys are labeled accordingly and important archaeological sites include Trapper Creek Overlook (TCO), Susitna River Overlook (SRO), Bull River II (BR), Carlo Creek (CR), Jay Creek (JC), and the Tangle Lakes (TL). Denali National Park is outlined in green. The Wisconsin glacial events are after Briner and Kaufman (2008) and Kaufman and Manley (2004). Map modified from AlaskaPak for ArcGIS 9.2 publically available from the National Park Service. See Figure 1.2 for additional references.

A series of post-LGM ice advances were less severe and represent the final gasps

of the late Pleistocene glaciation. These included the MP III at 15.1 ka and the MP IV,

coincident with the Younger Dryas from 12.3 to 11 ka (Briner and Kaufman 2008:663).

By the time humans apparently arrived in eastern Beringia ca. 14.5 to 14 ka the extent of

glaciation was diminished and a succession of climate and vegetation changes was

underway. These events were central to human adaptation as people slowly spread

throughout the state and toward the coast.

Dortch (2006) demonstrated glacial cycles in the Nenana valley corresponded to those in Denali National Park. Using terrestrial cosmogenic nuclides and optically stimulated luminescence dating techniques on glacial erratics, Dortch (2006) isolated the age of the Carlo advance between 19 and 17.2 ka, roughly equal to the MP II stadial.

Late Pleistocene erratics date the melting of localized glaciers at the summit of the

Reindeer Hills just east of Cantwell between 16.6 and 15.5 ka, and between 19 and 17.3 ka in the lower valleys of this region (Briner and Kaufman 2008; Dortch 2006). Less extensive glacial readvances in the Broad Pass area occurred during the Younger Dryas

12 ka (Dortch 2006) and probably would not have hindered human migration between interior and southcentral Alaska.

Southcentral Alaska. Additional glacial chronologies have been developed south of Trapper Creek, where a date of 10.4 ka (9200 BP) was obtained at the mouth of the

Little Susitna River’s upper canyon, providing a minimum date for ice-retreat in the mountains north of Palmer (Reger et al. 1995:40). A terminal moraine in Bootlegger’s

Cove in north Anchorage dates the Elmendorf stade of the Naptowne glaciation to 15 ka.

The Elmendorf glacier formed from merging ice flowing down the Matanuska-Knik valley from the Chugach Mountains, and via Turnagain Arm from the Kenai Mountains.

Immediately west of Bootlegger’s Cove, the lower Susitna River valley was isostatically depressed and flooded with glacioestuarine waters extending upstream during the

Elmendorf stadial (Reger et al. 2007:10). The channel of the middle and lower Susitna

River would have been significantly wider during the Elmendorf than at present as it moved a massive volume of glacial meltwater away from southcentral Alaska. Currently, there is no published evidence that the middle and lower Susitna valley were glaciated during the Elmendorf. Because glaciers retreated first from valley floors toward their alpine sources, it is not unreasonable to surmise that the middle Susitna River lowlands became ice-free shortly after the LGM and were not reglaciated. The question remains, were post-LGM ice advances severe enough to bring ice flows from the mountain valleys to Trapper Creek? Given the current lack of research between Cook Inlet and Broad

Pass, for now this question must remain unanswered, but the issue is revisited in Chapter

III.

The Aleutians and Southeast Alaska. Evidence from Cold Bay on the Alaska

Peninsula indicates that glacial ice centered over the Pacific shelf rapidly retreated between 16.7 and 13.9 ka (14,000 and 12,000 BP) (Mann and Peteet 1994). A willow leaf sample recovered from the bottom of a deeply stratified profile provided evidence for when the region became habitable, at least by vegetation. A radiocarbon assay from this leaf suggests vegetation was present 13.4 ka (11,520 ± 100 BP) (Jordan 2001:512).

Shortly after ice retreated, regional landmasses began rebounding as the immense weight of glacial ice was lifted. Isostatic uplift in Glacier Bay, southeast Alaska, was outpaced by worldwide increases in oceanic sea levels, which submerged any potential sites along the coast older than 15.4 ka (13,000 BP) (Mann and Hamilton 1995). Similar circumstances across the southern Alaskan coast make it difficult to understand the extent of, or potential for, terminal Pleistocene maritime economies. Some archeologists

(Yesner 1998, 2001) suggest that Alaskan maritime economies emerged as environmental

conditions in the interior deteriorated, i.e., changing paleoenvironments had a “push-pull”

effect on early peoples living in eastern Beringia that attracted foragers toward resource-

rich coastal zones. Others (Dumond 2001, Workman and McCartney 1998) add that site preservation biases in southern Alaska make the discovery of sites dated from 10.9 to 8.3

ka (9600 to 7500 BP) difficult. Poor site preservation is the result of complex coastal

interactions of eustatic, isostatic, and tectonic processes that eroded or obscured

coastlines from that period (Jordan 2001:520).

Climatic Oscillations

Several climatic shifts documented in eastern Beringia correlate with worldwide trends

that are relevant here (Table 2.1). The earliest among these are 1) the Bølling from 15.4

to 13.9 ka; 2) Older Dryas, a brief cold episode within the Bølling interval and dated in

some regions of North America from 14.2 to 14 ka; 3) a warm trend equivalent to the

European Allerød dated between 14 and 12.8 ka; 4) the Younger Dryas cold snap

between 12.8 and 11.5 ka, and 4) the Preboreal that produced a parkland similar to that of

northern Europe (Ager 1983, 1989; Björck 1996; Hall et al. 2004:133; Hendy et al.

2002).

Pinpointing precise time intervals for the onset of these events has been

problematic because worldwide climatic changes may have occurred differentially across

microenvironments within Beringia (Bigelow 1997; Bigelow and Edwards 2001;

Edwards et al. 2001; Elias 2001; Hoffecker and Elias 2003; Mann and Hamilton 1995;

Mann et al. 1998; Mann et al. 2001; Mann et al. 2002; Mason et al. 2001). For example,

late Pleistocene worldwide temperature increases lagged in some heavily glaciated

regions (Kaufman et al. 2004). Despite regional variations, many changes in climate and

ecology began rapidly worldwide and therefore are detectable in global proxy indicators

such as newly developed U/Th chronology from the combined Greenland ice core (GRIP

and GISP2) and speleothem records from the deeply stratified Hulu cave deposits in

China, which provide a high resolution record of global paleoclimate change (Weninger and Jöris 2008).

One crucial proxy record employed in correlating the GISP2-HULU data on a local scale (Fig. 2.2) is moraine features that document expanding and contracting glaciers, as well as pollen cores which can quantify changing frequencies of cold and warm loving plant species through time. Pollen analyses can aid in reconstructing prehistoric vegetation regimes, such as studies from Windmill Lake in Broad Pass that indicate a sparse herb tundra existed by 16.9 ka (ca. 14,000 BP) during cold and arid conditions and was replaced with a shrub tundra by 13.8 ka (ca. 11,800 BP) suggesting warmer climate and a mostly ice-free central Alaska at that time (Bigelow and Edwards

2001).

The Younger Dryas, a severe cold episode experienced more-or-less throughout the northern hemisphere, was a significant climate shift generally demarking the end of the Pleistocene epoch. Haynes (2002:267) places the event between 12.8 and 11.9 ka

(10,900-10,200 BP) in North America. However, the cold spell was not contemporaneous across North America. Mann et al. (2002) reported evidence for the

Younger Dryas in the northern Arctic foothills from 12.9 to 11.5 ka (11,000 to 10,000

BP). Bigelow and Edwards (2001:203) show vegetation shifts in Broad Pass consistent with the Younger Dryas between 12.5 and 11.9 ka (10,500 and 10,200 BP). Powers and

Hoffecker (1989:267) reported on the post-Carlo glacial advance in the Nenana valley at

approximately 12.5 ka (10,500 BP), which is coeval with an Artemisia spike in pollen

percentages and low lake levels in Broad Pass (Bigelow and Edwards 2001). Brubaker et

al. (2001) and Elias (2001:10) date the Younger Dryas onset to 12.7 ka terminating

abruptly at 11.5 ka (10,800 to 10,000 BP). However, as mentioned, problems with the

radiocarbon method exist from this period (Alley et al. 1993; Fiedel 1999:96; Waters and

Stafford 2007).

Figure 2.2. GISP2 HULU Delta 18O isotope climate curve after Weninger and Jöris (2008) and central Alaskan climate proxies. Notations as follow: Last Glacial Maximum (LGM), McKinley Park stadial (MP), Older Dryas (OD), and the Milankovitch Thermal Maximum (MTM). Swan Point (SP) CZ 4 represents the earliest known human presence in eastern Beringia (Holmes in press). See text for additional references.

Following the Younger Dryas, the Milankovitch Thermal Maximum (MTM) dates

between 10 and 9 ka (8900 and 8100 BP), a time when Alaska’s interior experienced a

warm, moist climate and the initial expansion of boreal forests (Bigelow 1997:iii;

Bigelow and Edwards 2001; Edwards and Barker 1994; Elias 2001:10-11; Mason et al.

2001:525). The MTM was followed by another cold trend, called the “younger-Younger

Dryas” by Mason et al. (2001:525, 538) and the “Mesoglacial” by Begét (1983:389).

Cooler temperatures during this time were sufficient to cause glacial advances and lower than normal lake levels across interior Alaska from 8.5 to 8 ka (7800 to 7200 BP).

Warming and cooling oscillations continued throughout the Holocene with at least five neoglacial advances documented at the Dry Creek site in the Nenana valley (Powers and Hoffecker 1989:267), the earliest dating to approximately 6.8 ka (6000 BP). The remaining cold episodes noted at Dry Creek could be associated with globally documented events dated between 4.5 and 4.2 ka (ca. 3800 BP), 3.1 to 2.1 ka (ca. 2900 and 2100 BP), and with a cooling event in the late Holocene (Brigham-Grette 2001:20) known as the Medieval glacial advance from 1.45 to 1.2 ka in northern Europe (ca. 1500 to 1300 BP) (Calkin et al. 2001:449).

Paleoecology of the Mammoth Steppe

A secondary consequence of increased atmospheric aridity during late Pleistocene stadials was the expansion of a tri-continental grassland described as the mammoth steppe (Guthrie 1990, 2001, 2006), a biome that spread from central Asia to northeastern

Europe in the west and the Yukon Territory in the east. The mammoth steppe was a result of extensive ice sheets that trapped or redirected moisture from the east and west in a combination of cold and arid conditions that diminished woody plants. Fertile soils favored highly nutritious hardy grasses and forbs that were easily digestible by grazers.

In eastern Beringia, the forage played host to a plethora of megafauna grazers including

steppe bison (Bison priscus), horse (Equus sp.), reindeer (Rangifer tarandus), muskoxen

(Ovibos moschatus), Saiga antelope (Saiga tartarica), camel (Camelops sp.), and woolly mammoth (Mammuthus primigenius), among others. It was the quintessential open range with frequently clear but probably dusty skies and little rain (Guthrie 1990, 2001).

However, this large expanse did contain patches of woody plants and shrubs in low lying areas, and alpine tundra at higher elevations. Consequently, it also supported small communities of browsers. Thus the “mammoth steppe” supported diverse fauna (Graf

2008; Guthrie 2001).

The mammoth steppe undoubtedly fostered the expansion of modern humans into the Arctic and eventually interior Alaska (Graf 2008; Guthrie 1990). However, the first human groups arrived in eastern Beringia during the turbulent terminal Pleistocene (16 to

11 ka) when a rapidly declining mammoth steppe forced late Pleistocene megafauna into refugia within a 30 km stretch of foothills bordering the northcentral Alaska Range (Ager

1975:85-86; Powers and Hoffecker 1989; Yesner 1996). The Front Range refugia offered stable footing and nutrient-rich grasses at a time when most of interior Alaska was transitioning to the highly acidic and nutrient-poor tussocks found across the region today (Walker et al. 2001).

The termination of the Pleistocene epoch marked the end of the latest glacial period around 12 ka (10,000 BP) in eastern Beringia when annual average temperatures increased and vegetation landscapes were drastically altered (Ager 1983; Bigelow 1997;

Guthrie 1983; Kaufman et al. 2004:536). Climate change at that time was severe, and combined with the introduction of human predators, most Pleistocene megafauna were stressed to extinction (Alroy 2001a, 2001b; Brook and Bowman 2002, 2004; Guthrie

1990, 2006; Koch and Barnosky 2006; Koch et al. 1998; Koch et al. 2004; Martin 1984;

Mosimann and Martin 1975; Whittington and Dyke 1984; Winterhalder et al. 1988). In

fact, evidence suggesting that humans may have hunted mammoth or horse species in

eastern Beringia is sparse, found in only one small assemblage at Swan Point CZ 4 dated

to 14.3 ka (Holmes in press).

Although the role of human hunting in the extinction of steppe fauna is still

debated, the selective die-off of megafauna in Alaska would have also strained

Paleolithic hunting economies adapted to the mammoth steppe. Ecological change at the

Pleistocene-Holocene transition would have forced foraging societies to re-adapt, and

even migrate as they struggled to invent new subsistence strategies almost immediately

upon entering the New World. A series of terminal Pleistocene climatic oscillations, the

Bølling, Older Dryas, Allerød, and Younger Dryas, between ca. 15.4 and 11.7 ka slowly

phased the mammoth steppe into the Preboreal period.

Bison inhabited the eastern Beringian mammoth steppe long before the LGM, and

the genus survived the selective megafauna extinction through a series of evolutionary

shifts and by transitioning to the parkland/woodland (Guthrie 2006). The encroachment

of these species into partially forested regions was possible in the late glacial with the

initial spread of the northern forests from southern Alaska during the Allerød, resulting in

a patchwork of trees and steppe best described as parkland.

Less is known about the Preboreal period in Alaska, only that it is associated with

the northern spread of forests that bordered the southern margins of the open mammoth

steppe during the Bølling/Allerød. The term Preboreal was originally applied in northern

Europe, but there is an Alaskan analogue equivalent to Subzone WL-2b defined by

Bigelow and Edwards (2001:208) in the Windmill Lake pollen record from Broad Pass in

central Alaska. The Preboreal stage is characterized by the initial spike in Betula (birch)

and Salix (willow) followed by the decline of those species at the expense of Populus sp.

(aspen and cottonwood), and Artemisia (shrubs and herbs) between 11.1 and 9.4 ka (9600

to 8300 BP). The development of the Preboreal in eastern Beringia signaled the end of

the productive mammoth steppe environment envisioned by Guthrie (1990).

In Asia and parts of Europe, the Preboreal was host to sizeable communities of

browsers and mixed feeders including wapiti (Cervus canadensis), red deer (Cervus

elaphus) and roe deer (Capreolus capreolus), wild boar (Sus scrofa), and moose (Alces

alces) with a similar array of woodland species thriving in southern and central Alaska

during this time (Guthrie 2006). Prehistoric foragers frequently procured a number of

these species and supplemented their diet with a variety of fish (salmonids) and birds

including swan (Cygnus sp.) and dabbling ducks (Anas sp.) during the Pleistocene-

Holocene transition, as indicated by faunal assemblages from Dry Creek I and II (Powers

and Hoffecker 1989), Broken Mammoth CZ 3 and 4 (Yesner 1996, 2001), and Swan

Point CZ 4 (Holmes 1996, in press). Another important game species hunted during the

Pleistocene/Holocene transition was Dall sheep (Ovis dalli). Genetic studies indicate that sheep probably survived the LGM in interior Alaska and migrated into the central Alaska

Range behind the retreating ice where they survive today (Loehr et al. 2006). If the number of game animals in the central Alaska Range at the Pleistocene-Holocene boundary (13-10 ka) resembled the substantial concentration of game reported at the time of Euro-American contact (Walker 2005), then it is by no coincidence that the earliest sites in Alaska occurred within or near the Front Range (Ten Brink and Waythomas

1985; Yesner 2001:316).

Not all of these climatic oscillations were detrimental to (and some may have

even provided opportunities for) early foragers. The Bølling/Allerød saw increases in

human populations as climates ameliorated, but the Younger Dryas cold snap at 12.7 ka

had the opposite effect, nearly forcing small-scale foraging societies to share the fate of the mammoth (Chapter V). It is not until the Younger Dryas that the first signs of

humans appear south of the central Alaska Range, at sites like Bull River II (Chapter IV;

Wygal 2007c), Carlo Creek (Bowers and Reuther 2008), and potentially in the Tangle

Lakes region (West et al. 1996a, 1996b, 1996c). This attraction was likely the result of increasingly harsh conditions in the interior (Yesner 1996) and improving situations in southcentral Alaska with the still developing Preboreal ecosystem.

The Boreal stage occurs in two parts (Table 2.1). In its earliest phase, coincident with Pollen zone WL-3 and WL-4 from the Windmill Lake core (Bigelow and Edwards

2001:208), it was likely a productive environment, with stands of birch (Betula sp.), white spruce (Picea glauca) and alder thickets (Alnus sp.) shortly after 9.4 ka. Although the Boreal was probably less productive than the Preboreal period, productivity declined markedly in southcentral Alaska as conditions became increasingly mesic ca. 6 ka and black spruce (Picea mariana) expanded quickly across the region (Guthrie 1990). One ramification of black spruce emerging as the dominant woody plant species in the boreal forest is a consequential increase in the frequency and intensity of natural forest fires across southcentral Alaska (Lynch et al. 2002).

The source of these woody plant species, primarily Populus, Larix, Picea, Betula,

Pinus, and Alnus, which formed the boreal and preboreal forests in different percentages since the LGM, is still debated. One explanation proposes a rapid invasion of these

species into Beringia from latitudes south of the Canadian ice sheets (Guthrie 1990).

Alternatively, a glacial refugium model maintains these species survived throughout the

LGM in small communities across Beringia, spreading from small holdout colonies

during ameliorating climates. Evidence is gaining to support the latter explanation which

depicts a mosaic of plant and animal species in a patchwork of ecological niches across

the mammoth steppe during and following the LGM (Anderson et al. 2006; Brubaker et

al. 2005). A similar environment was described by Graf (2008) for western Beringia,

suggesting the first Alaskans were preadapted to the conditions in eastern Beringia when

they arrived.

The boreal forest refugium hypothesis advanced by Brubaker et al. (2005)

explains the arrival of moose in eastern Beringia. Genetic studies of global moose populations suggest a genesis in Eurasia and arrival in eastern Beringia via the land bridge (Hundertmark et al. 2002; Kurten and Anderson 1980) contemporaneous with humans ca. 14.3 ka (12,201 ± 85 BP) (Guthrie 2006 supplementary data). Obligate browsers, moose began to thrive during the Preboreal and Boreal periods. Although bison and elk are grazers, their digestive systems allow for a greater mix of forage, including willow and twigs. This ability helped them survive the Preboreal, unlike horse and mammoth which did not fare as well (Guthrie 1990, 2006).

In the middle Susitna valley, vegetation likely was comprised of a mixed steppe- tundra environment immediately following deglaciation. Prior to the Preboreal period and before seasonal salmon runs were established, early foragers may have focused on remnant grazers such as bison, and potentially the occasional mammoth, near the shores of the Susitna and Chulitna Rivers (Bacon et al. 1983:12-13). Early reconstructions of

the Susitna lowland ecosystem depict a slow transition toward a parkland during the

Preboreal with scattered stands of woods, and eventually an established aspen/birch forest by 10 ka (9000 BP) (Ager 1975; Bacon et al. 1983:10). However, these depictions were based on generalizations from only a few primary data sources in the 1970s and 80s.

A recent review of pollen core data from across eastern Beringia indicated woody plant species began spreading throughout southcentral Alaska in the late Pleistocene, as much as 6,000 years earlier than previously demonstrated (Brubaker et al. 2005). Betula pollen was present in high percentages (>60%) around upper Cook Inlet as early as 16 ka and dominated southcentral Alaska by 13 ka. Populus appears in low frequencies (<2%) at 14 ka and increased to >10% in some areas by 11 ka, declining to lower levels by 9 ka.

Alnus appears in low percentages (<10%) by 14 ka while Picea arrives in the area in small amounts (<2%) 14 ka at one core location on the northern Alaska Peninsula. It appears in similar frequencies across southcentral Alaska by 12 ka (Brubaker et al. 2005).

Although this setting was ideal for browsers such as moose, and browser/grazers such as wapiti and bison, eventually a sharp rise in Picea, particularly black spruce, at the expense of hardwoods and grasses ca. 6 ka increased plant toxicity and reduced foraging productivity (Guthrie 1990). It was during this period that salmon runs probably emerged as an important component of the southcentral Alaskan ecosystem (Bacon et al. 1983;

Matson and Coupland 1995) and with it, the abandonment of big game hunting as a mainstay.

Today the ecology of the Susitna River lowlands has changed little from 6 ka.

Modern vegetation in the region consists primarily of a mixed black spruce, white spruce, and birch (Picea and Betula) forest. Small stands of aspen and cottonwood (Populus)

and Alaskan larch (Larix laricina) trees are present but rare. The undergrowth consists of

Fiddlehead fern (Matteuccia sp.) and Devil’s club (Oplopanax horridus) making many areas impassible by foot in mid-July. In early spring, fern sprouts and Devil’s club buds are harvested and consumed by people living in the area today. Mixed among the dense forest are intermittent streams, peat bogs, kettle ponds, and swamps, typically bordered by small colonies of tussocks or muskeg (Eriophorum sp.). Major river channels are lined with substantial stands of alder thickets. Local animal populations include moose

(Alces alces), black and brown bear (Ursus americanus and U. arctos), wolf (Canis lupus), and to a lesser extent porcupine (Erethizon dorsatum) and ptarmigan (Lagopus sp.). Fish include arctic grayling, river trout, northern pike and seasonal runs of king, silver, and chum salmon.

Volcanism

Positioned at the north crest of the Pacific Ring of Fire, Alaska had and continues to have a rich history of volcanic activity, particularly along its southern margin and especially on the Alaska Peninsula and throughout the Aleutian Islands. Documentation of volcanic events across southern Alaska has two significant functions: 1) prominent tephra deposits serve as regional chrono-stratigraphic indicators (Begét et al. 1991; Bowers 1979; Child et al. 1998; Dixon and Smith 1990; Lemke 2000; Riehle 1994; Riehle et al. 1990; Riehle et al. 1998; Schiff et al. 2008:66); and 2) increases in extensive eruptions between 4.5 and

3.6 ka (4000 and 3400 BP) likely influenced ecology and human behavior in southern

Alaska (VanderHoek and Myron 2004). Two well documented tephra horizons are prominent throughout southern and central Alaska (Fig. 2.3) and cap archeological deposits in many sites in southcentral Alaska (Bowers 1979; Dixon and Smith 1990;

Wygal and Goebel 2006). Of these, the Hayes set H tephra is the best known (Begét et

al. 1991; Riehle et al. 1990; Riehle 1994; Schiff et al. 2008:66). Less is known about the

Oshetna tephra, including its source vent, although, because of similarities in mineral

signatures, the Hayes vent west of Anchorage is the suspected vent for the Oshetna tephra

(Child et al. 1998). Dates on Oshetna deposits had not been well established but estimated to have been between 6.8 and 5.9 ka (6750-5850 cal BP) according to some reports (Dixon 1993; Dixon and Smith 1990), but recently Child et al. (1998) dated the deposit in Denali National Park to ca. 6 ka. The distribution of the Oshetna tephra was further expanded with its recent discovery at Bear Lake at the base of Redoubt Volcano along the western shores of Cook Inlet west of Anchorage (Schiff et al. 2008).

The Hayes tephra represents the most extensive tephra deposit across southcentral and central Alaska, but it did not occur as a single eruption. Instead it was deposited in a series of eruptions dated between 4.3 and 3.8 ka (ca. 4200-3800 cal BP). It has been independently dated in southcentral Alaska to 4.1 ka (4000 ± 210 cal BP / 3650 ± 150

BP) (Begét et al. 1991; Riehle et al. 1990). The Hayes tephra has three localized manifestations including the Cantwell ash from the Carlo Creek site (Bowers 1979), the

Jarvis Creek ash near Delta Junction (Péwé 1975), and the Tangle Lakes ash (Begét et al.

1991). It has recently been found in cores from Wonder Lake and Sneaker Pond north of the Alaska Range in Denali National Park where a maximum bracketing age of 4.3 ka

(4250 ± 100 cal BP / 3830 ± 60 BP) was obtained (Child et al. 1998). Cores from Bear

Lake also contain the deposit where Schiff et al. (2008) obtained an age of 4.1 ka (4030 ±

90 cal BP) from the base of the ash. Dates from the Kenai Peninsula are consistent and help demonstrate the widespread nature of these deposits (Schiff et al. 2008).

Figure 2.3. Known distribution of the Hayes set H and Oshetna tephra deposits. The Hayes tephra deposits originated from the Hayes Volcano west of Cook Inlet in a series of eruptions dated between 4200 and 3800 cal BP. Distributions mapped here were modified from Begét et al. (1991), Dixon and Smith (1990), Péwé (1975), Riehle (1994), and Riehle et al. (1990), and incorporate recent finds by Child et al. (1998) and Schiff et al. (2008). Map modified from AlaskaPak for ArcGIS 9.2 publically available from the National Park Service.

Primary Evidence for Volcanism in the Trapper Creek Locality. To better correlate known regional tephra deposits with localized tephrochronological markers in the study region (Trapper Creek to Carlo Creek) as well as understand the post-glacial ecology of the Susitna River lowlands, a peat bog was cored in cooperation with

Randolph Tedor (2006), a University of Alaska, Anchorage student in geoarcheology, and the 2005 and 2006 University of Nevada, Reno archeological fieldschools. An isolated kettle pond with no inlet or outlet streams and no sign of beaver den or dam activity was selected to minimize fluvial and bioturbation disturbances. The pond is located 3.4 km north of the Trapper Creek community and less than 1 km north of the

TCO archeological site (see Fig. 3.1). Thus, its chronology is an accurate depiction of

Holocene ecological conditions at TCO and SRO. Three ash horizons (Tephras 1 – 3) were identified in the 2005 core (Fig. 2.4). The Tephra 2 ash package (Section B, 107-

131 cm) was dated between 3.6 and 4.25 ka (3577 ± 41 and 4190 ± 41 BP / 3880 ± 60 and 4730 ± 60 cal BP), on bracketing macrobotanical remains, contemporaneous with the

Hayes set H tephra. A third date of 8331 ± 41 BP (9360 ± 60 cal BP) came from a thin but densely packed horizon of macrobotanical remains (Section C, 226 cm) corresponding to the initial rise of birch forests in the Trapper Creek area (Tedor 2006).

Figure 2.4. Trapper Creek bog 2005 core. Sedimentology after Tedor (2006).

Table 2.2. Summary of electron microprobe analysis, Trapper Creek bog 2005 core (Wan 2007).

Tephra 2/119-121 cm (combo) Sample Na2O MgO Al2O3 SiO2 K2O CaO TiO2 MnO FeO Total Mean 2.981 0.429 14.003 72.692 2.483 1.911 0.201 0.071 1.453 96.222 Standard Deviation 0.797 0.047 0.488 2.086 0.126 0.115 0.042 0.027 0.110 2.620 Range 2.404 0.146 1.707 6.887 0.425 0.357 0.116 0.087 0.384 9.245 Minimum 1.398 0.368 12.750 67.876 2.211 1.769 0.160 0.032 1.309 89.025 Maximum 3.802 0.514 14.457 74.763 2.636 2.126 0.276 0.119 1.693 98.270 Count 10 10 10 10 10 10 10 10 10 10 Confidence (95.0%) 0.570 0.033 0.349 1.492 0.090 0.082 0.030 0.020 0.079 1.874 Tephra 2/127-128 cm Mean 4.287 0.087 29.239 54.521 0.283 11.255 0.050 0.017 0.541 100.279 Standard Deviation 0.429 0.151 4.307 5.322 0.520 2.713 0.064 0.017 0.403 0.785 Range 1.792 0.512 17.706 21.619 2.049 11.488 0.214 0.063 1.347 3.542 Minimum 3.542 0.010 14.283 51.285 0.068 2.051 0.000 0.000 0.299 97.955 Maximum 5.334 0.522 31.989 72.904 2.117 13.539 0.214 0.063 1.646 101.497 Count 17 17 17 17 17 17 17 17 17 17 Confidence (95.0%) 0.221 0.078 2.215 2.736 0.267 1.395 0.033 0.009 0.207 0.404 Tephra 3/135 cm (Run 1) Mean 4.237 0.039 30.707 52.578 0.111 12.196 0.027 0.009 0.413 100.315 Standard Deviation 0.286 0.011 0.448 0.761 0.024 0.646 0.019 0.009 0.047 0.464 Range 1.068 0.031 1.443 2.807 0.113 2.223 0.058 0.027 0.205 1.537 Minimum 3.811 0.023 29.837 51.533 0.079 10.878 0.000 0.000 0.310 99.607 Maximum 4.879 0.054 31.280 54.340 0.192 13.101 0.058 0.027 0.515 101.144 Count 20 20 20 20 20 20 20 20 20 20 Confidence (95.0%) 0.134 0.005 0.210 0.356 0.011 0.302 0.009 0.004 0.022 0.217 Tephra 3/135 cm (Run 2) Mean 5.034 0.040 30.139 53.404 0.129 11.875 0.018 0.005 0.403 101.046 Standard Deviation 0.773 0.019 1.040 1.849 0.030 1.274 0.014 0.008 0.109 0.586 Range 2.580 0.060 3.559 6.099 0.111 4.295 0.037 0.027 0.395 2.274 Minimum 4.148 0.010 27.862 51.328 0.088 9.022 0.000 0.000 0.151 99.704 Maximum 6.728 0.070 31.421 57.427 0.199 13.317 0.037 0.027 0.546 101.978 Count 18 18 18 18 18 18 18 18 18 18 Confidence (95.0%) 0.384 0.010 0.517 0.919 0.015 0.633 0.007 0.004 0.054 0.291 Jarvis and Oshetna Reference Samples (from Childe 1998: Table 1) Jarvis Creek Mean 3.93 0.54 14.54 73.67 2.59 2.25 0.24 ― 1.86 100 Count 11 11 11 11 11 11 11 ― 11 11 Standard Deviation 0.11 0.07 0.12 0.53 0.07 0.15 0.04 ― 0.21 0 Oshetna Mean 4.20 0.66 14.63 72.55 2.72 2.31 0.43 ― 2.29 100 Count 22 22 22 22 22 22 22 ― 22 22 Standard Deviation 0.36 0.36 .70 1.00 0.17 0.32 0.10 ― 0.42 0

Tephra samples from the Trapper Creek bog were mineralogically characterized using an electronic microprobe technique at the USGS tephrochronology laboratory in

Menlo Park, California (Wan 2007). These results are compared in Table 2.2 to

reference samples from the Jarvis and Oshetna tephras published in Childe et al. (1998); however, heavy weathering of glass shards in the Trapper Creek bog tephra caused degradation of the sample into crystalline and/or clay, a common problem in the andesitic to dacitic based tephras from Alaska. Unfortunately, these processes resulted in poor to variable analytical signatures from the Trapper Creek sample and hindered positive correlations with others in the region (Wan 2007). Nevertheless, the mineralogical signatures of the samples may still prove useful (Wan 2007) particularly in later discussions on stratigraphy and dating of TCO and SRO (Chapter III).

Conclusions

The Bering Land Bridge was the ancient link between the Old and New Worlds during the late Pleistocene and allowed the migration of plants, animals, and eventually people.

Because the continental shelf is relatively shallow through the Bering Strait, a worldwide decrease in sea levels during the LGM caused the exposure of the land bridge.

Continental ice sheets over northern Europe and most of Canada and the continental

United States blocked currents of atmospheric moisture and increased the aridity of northern latitudes across the globe. Compared to saturated tundra, swamps, and boreal forests found in northern latitudes today, during the late glacial the north received considerably less rain and cloud cover. The aridity fostered the expansion of the mammoth steppe, a mosaic biome of extended steppe-tundra, with small colonies of shrubby and woody plants, particularly along its southern margins. The mammoth steppe supported large herds of megafauna and enabled the expansion of Paleolithic hunters across the steppe, and over the land bridge. Reconstructions of the paleoecology, glacial cycles, and volcanism in central and southcentral Alaska establish the context in which

prehistoric foragers were constrained by and/or benefited from changes in their environment.

At first, the continued migration of highly mobile hunter-gatherers further south into Canada was blocked by colossal piedmont glaciers. As the climate warmed during the late glacial period, ice reduction was rapid and vegetation recovery in the Susitna valley and upper Cook Inlet was swift, with woody plants, shrubs and grasses expanding their range north between 16 and 14 ka. Although these conditions proved too drastic for mega-grazers including mammoth and horse, it was particularly favorable to browsers and browser/grazers including moose, bison and wapiti. Prehistoric foragers took advantage of these resources by expanding their geographic range once glaciers receded, and by following prey species. As areas north of the Alaska Range deteriorated further during the Preboreal stage, the Front Range and southcentral Alaska were probably quite habitable and teeming with medium sized game species in alpine refugia at higher elevations and a parkland environment in the broad valleys. The Susitna River valley consisted of open meadows and stands of hardwood forests in what would have been an active niche for small scale foraging societies. However, the productivity in southern

Alaska was short-lived with increases in volcanism and the eventual development of a relatively unproductive mesic boreal forest ecosystem during the middle Holocene.

Human diasporas throughout eastern Beringia were influenced by these factors, pushing and pulling them between environments of varying degrees of productivity across the state. As terrestrial resources declined in the middle Holocene, salmon runs emerged and coastlines stabilized, forcing an economic shift to marine and riverine resources, including salmon in the interior and sea mammals along the coast.

CHAPTER III

ARCHEOLOGY OF THE SUSITNA RIVER LOWLANDS

While southcentral Alaska remains relatively unknown archeologically, particularly with respect to its earliest occupants (Wygal and Goebel 2006), several cornerstone investigations have established a basic chronology for the major geographic sub-regions including the area south of Broad Pass, Tangle Lakes, Susitna River basin, Alaska

Peninsula, and Kodiak Island (Bacon et al. 1983; Bowers 1978, 1979, 1980; Clark 1998;

Dixon et al. 1985; Dumond 1975, 1998; Henn 1979; Jordan 1992; Reger 1981, 1996,

1998; West 1996a, 1996b, 1996c). Despite the existing dataset, southcentral Alaska is immense with much of the region being inaccessible to archeologists. Past survey reports have emphasized the difficulty in finding buried and datable archeological deposits

(Bacon 1983; Bacon et al. 1983; Dixon et al. 1985). Physiographically, the region is dominated by the Susitna River and its tributaries, which drain the uplands of the Alaska and Talkeetna mountain ranges south to Cook Inlet. Peaks in these ranges reach 6000 m with glaciers in many upland valleys and unglaciated landscapes that are mantled by thin soil horizons and tundra plant communities. Sites have been discovered and documented in these contexts, but most are surface or near-surface sites that have not been reliably dated (Chapter I). In the Susitna lowlands, landscapes are conversely mantled by stratified loess and tephra deposits, sometimes reaching 1 m or more in thickness, but the area is densely vegetated with boreal forest, making surface visibility, let alone pedestrian survey, difficult and sometimes perilous.

The research presented here details the discovery and excavation of two buried and stratified archeological sites located near the confluence of the Susitna, Chulitna, and

Talkeetna rivers, in southcentral Alaska. I became interested in this area because of its

strategic position between the interior of Alaska, home of numerous terminal Pleistocene

and early Holocene archeological sites, and the maritime environments of Cook Inlet and

the Gulf of Alaska. The Nenana River valley is located due north of Trapper Creek, and flows to the Tanana River just to the north of Broad Pass, one of the lowest passes through the Alaska Range. If human populations dispersed from the interior to coastal regions during the terminal Pleistocene or early Holocene as Yesner (1998; 2001:316) and others (Ackerman 1992; Bacon et al. 1983; Dumond 1998:190; Moss 1998; Reger

1981, 1998) have suggested, then evidence of this event should be found in the archeological record of the Susitna Valley. Questions guiding this research include: (1)

When did humans first appear south of the Alaska Range? (2) How did early humans use the mountainous and riverine environments of southcentral Alaska—was there a lengthy period of exploration and seasonal use or did a resident population of humans quickly settle into the Susitna lowlands? (3) How did the early users of the region adapt technology, subsistence, and settlement to meet the requirements of this unique northern environment?

This chapter presents the results of research at the Trapper Creek Overlook (TCO) and Susitna River Overlook (SRO) archeological sites, including site discovery, geomorphic and stratigraphic contexts, chronology, and archeological contents. Because faunal remains are not preserved at the sites, the record primarily consists of lithic artifacts. These data have important implications for understanding the early archeological record of central and southcentral Alaska, and for reconstructing technological organization and settlement during the early Holocene.

Study Area

The SRO (TAL-095) and TCO (TAL-092) sites are located west of the Susitna River, near its confluence with two other major southcentral Alaskan rivers, the Chulitna and

Talkeetna. In the vicinity of the sites, these rivers flow across the broad alluvial plain of the Susitna Valley, but less than 20 km northeast of the sites the three rivers exit relatively narrow, steeply incised canyons. The Susitna’s alluvial plain is nearly 40 km wide around modern communities of Trapper Creek and Talkeetna. Both sites occupy elevated knolls that overlook the otherwise low-lying and flat terrain. TCO can be accessed via a short hike from the George Parks Highway (milepost 117), while SRO is more remote, requiring river boat access and a 2 km trek west through the forest.

SRO is located 12 km south of the town of Trapper Creek, while TCO is 3.5 km north of town. Situated on a prominent glacial esker, SRO overlooks a fossil slough of the Susitna River that is presently an elongated swamp (Fig. 3.1). Accessing the site would have been significantly easier when the fossil slough was active. The esker rises

22 m above the first terrace of the Susitna River, itself 25 m above the modern level of the river. This is the only obvious elevated landform in the immediate area.

TCO is also situated on a high glacial esker immediately east of and 20 m above a

series of kettle ponds, and Trapper Creek, a small tributary of the Susitna. From the site

one has a clear view of the surrounding floodplain as well as the tri-river convergence 4.5 km to the east. Today, the community of Talkeetna is clearly visible from this location despite the dense boreal forest that normally hinders visibility. Elevated features like those that contain the SRO and TCO sites are relatively infrequent along the first terrace

of the middle Susitna River; generally, the area is flat and in the past the landforms containing the sites could have functioned as strategic locations for prehistoric hunters.

The eskers that contain SRO and TCO were formed by a stagnated ice sheet most likely dated to the LGM ca. 22 to 19 ka; alternatively, they could be contemporaneous with the Elmendorf readvance dated to 15 ka in Bootlegger’s Cove north of Anchorage

(Reger et al. 2007). The most significant unanswered question regarding the glaciation of the Trapper Creek region surrounds the extent of the Elmendorf or its equivalent in the

Susitna River lowlands between Cook Inlet and Broad Pass. The Elmendorf formed from merging ice flowing down the Matanuska and Knik valleys out of the Chugach and from the Kenai Mountains via Turnagain Arm. Immediately west of Bootlegger’s Cove, the lower Susitna River valley was isostatically depressed and flooded with glacioestuarine waters extending upstream during the Elmendorf stade (Reger et al. 2007:10). At 15 ka, the Susitna River was significantly wider than at present as massive volumes of glacial meltwater channeled away from the mountains of southcentral Alaska. Additional research is required to determine whether glaciation in southcentral Alaska was limited to mountain valleys and foothills following the LGM or whether lower elevations and major river valleys contained ice mantles until the early Holocene. Understanding the glacial history and formation of the eskers that contain the sites is central to interpreting the post- glacial human colonization of the region.

Excavation History

In 2004, a team of five Matanuska-Susitna (Mat-Su) Borough archeologists, me included, under the principal investigator Fran Seager-Boss, discovered and tested the SRO and

TCO sites (Wygal and Seager-Boss 2005). At both sites, stone tools were recovered in

buried and stratigraphically sealed contexts. The sites were revisited in 2005 and 2006, when block excavations were undertaken. At SRO, an area of 18 m2 was excavated. At

TCO, the Mat-Su team excavated 7 m2 in 2004, and the UNR fieldschools completed a

total of 53 m2 in 2005 and 2006 (Fig. 3.1).

Figure 3.1. Susitna River lowlands with excavation history of the TCO and SRO sites. Modified aerial photo includes material by © Space Imaging LLC.

Both sites were carefully excavated to basal till, horizontally using 50 cm quadrants and vertically following natural stratigraphy and 5 cm arbitrary levels. At both sites artifact proveniences were recorded using metric tape-measures, plumb bobs, and line levels by excavators who also recorded the geologic stratum of the artifacts. Further

care was taken at TCO, where a centralized total station theodolite was used in addition

to these methods. A series of 50 cm shovel probes set on a 2.5 m grid across the site

were excavated to establish site boundaries and direct block excavations. All excavated

sediments were screened through 1/8-inch wire mesh, and screen finds were recorded by unit quadrant and 5-cm arbitrary level. Stratigraphic profiles and charcoal “features” were carefully documented and photographed. These datasets were later analyzed and compared to site stratigraphy and topography using Golden Surfer 8 software. The results allowed for the separation of two components at TCO based on a vertical distance of 5 to 7 cm between the upper and lower assemblages at the site.

Stratigraphy

Loess accumulation and paleosol formation provide information about paleo-

environments and glacial-interglacial cycles (Begét 1990; Edwards and McDowell 1991;

Hamilton et al. 1983; Manley et al 2001; Muhs et al. 2003; Péwé 1951, 1975; Reger et al.

1996; Westgate et al. 1990; Yesner et al. 1992), and these processes play a significant

role in understanding the geology of Trapper Creek. Loess deposits are swept by

katabatic winds from glacial outwash fans, deposited like a blanket on the landscape in thick sheets (Waters 1992:202). In Europe (Rapp and Hill 1998:174) and on the North

American Plains (Ruhe 1983) alternating layers of paleosol and unaltered loess are directly associated with glacial and interglacial periods. In interior Alaska, these alternating deposits are not as straightforward due to the number of glaciers active in

Alaska throughout the Holocene. Here, the process is viewed as a competing process

with pedogenesis occurring during periods of diminished loess accumulation associated

with interglacials when katabolic winds were reduced (Muhs et al. 2003:1973).

Methods

The following description of the sediments at TCO and SRO was developed over several

years in consultation with a variety of researchers. The first assessments and basic

stratigraphic profiles were documented in 2004 by Wygal, Krasinski, and Tedor which

led to the identification of seven individual strata including a prominent paleosol, upper

loess sequences, and the Hayes tephra marker, among others. In 2005, I used a soil guide

for field stratigraphic analysis based on the Unified Soil Classification System as well as

the US Department of Agriculture Soil Texturing Field Flow Chart (Midwest

Geosciences Group n.d.) to more accurately characterize the sediments at TCO and SRO.

These procedures were repeated in a more detailed analysis by Kelly Graf in 2006 at the

TCO site. Graf further refined our understanding of the stratigraphy by identifying

several additional layers including a second, less defined paleosol complex and an

additional very a fine tephra horizon. In 2007, a group of prominent Alaskan

archeologists and a geologist visited the site and offered their interpretations of the TCO

profile. These assessments largely agreed with the previous analyses in terms of strata

descriptions; however, it should be emphasized that there was skepticism by some in the

group over the unusual results of optically stimulated luminescence (OSL) dates from the

lower sequences of the profile. Nevertheless, that meeting resulted in the confirmation of yet another tephra horizon and confirmed that cryoturbation and other post-depositional disturbances did not seriously affect the observed profiles, which were generally representative of the two sites.

Stratigraphic Analysis

Aeolian deposits mantling the till at SRO and TCO reach more than 1 m in depth with

two paleosol complexes. The paleosols at TCO contained two cultural components,

while a single component was unearthed at SRO. Since the SRO and TCO sites have

nearly identical stratigraphic profiles, with all of the same stratigraphic units and horizons

represented, they are described here together. Details outlining differences between the

sites are specified below. Strata thickness, depths, and horizons are presented in Table

3.1. Contacts between most strata are quite distinct but some are gradual due to slow depositional rates. At least one of these gradual transitions has become blurred from solifluction, most notably in the lower layers of TCO component II (TCO-II), where it caused minimal post-depositional movement of artifacts. Slight upward drift probably caused by cryoturbation did move some artifacts in TCO component I (TCO-I) and SRO, but this disturbance was not severe, and based on a lack of refits between TCO-I and

TCO-II, did not lead to mixture between these components. Using thorough excavation and documentation techniques, these relatively minor post-depositional disturbances were easily identified and mitigated.

The basal layer (stratum 1) represents ablation till, a series of poorly sorted round to angular rocks and boulders ranging from pebbles to large cobbles. Rock types in this layer include schist, greywacke, mica, and quartzite; they are encased in a matrix of silt and sand. Stratum 2 is manifested differently between the sites; at TCO it appears as a thin (1-2 cm) lens of potentially fluvial coarse-grained sand that caps the till (Fig. 3.2).

The same deposit at SRO ranges to more than 10 cm in thickness and is mixed with higher amounts of silt. Stratum 2 is suspected to date to a period of post-glacial outwash.

Stratum 3 is a 25-to-35 cm thick deposit of loess that grades from a greenish-gray loamy

sand at its base (stratum 3a) into a yellowish-brown sandy loam at its top (stratum 3b).

Stratum 4 is a sandy clay loam that represents a second loess deposit, about 15-20 cm

thick. In its lower part (sub-stratum 4a) it contains discontinuous pockets of gleyed silty

clay loam and ash representing a reworked tephra (Graf 2006, field notes). Mineralogical analyses of the tephra suggest it suffered post-depositional leaching of its finer minerals, making it difficult to correlate chemically with other known Alaskan tephras (Wan 2007).

Figure 3.2. Stratigraphy and dating of TCO. The ages above include calibrated radiocarbon (italics), OSL samples (normal font), and dates on the Hayes tephra after Schiff et al. (2008).

Solifluction has blurred the contact between substrata 4b and 4a, but 4b is clearly

visible as a bright orange and compact horizon. At TCO this stratum contains high

concentrations of wood charcoal particularly on its upper surface where large clusters are

pervasive. Stratum 5 is a pale greenish-gray band of sandy loam reaching ca. 5 cm in

thickness. It contains a small amount of devitrified glass and shares a sharp contact with

stratum 6. The stratum 5 tephra probably represents the Oshetna tephra (Dixon and

Smith 1990) dated to ca. 6000 cal BP (Child et al. 1998); however, it is poorly preserved

in the Trapper Creek area making mineralogical characterizations unreliable.

Alternatively, stratum 5 may have been leached from the overlying stratum 6 deposits, a

1-2 cm thick bright yellow tephra with biotite phenoclasts that link it to the Hayes set H

tephra (K. Wallace 2007 pers. comm.).

Strata 7 and 8 are sandy loams (loesses) together reaching 25 cm thick. They are capped by strata 9 and 10, two light-gray tephras reaching about 1 cm thick each. In

some places across the sites, the strata

9 and 10 tephras were separated by a

thin layer of loess with an undulating

upper contact; in other areas the two

tephras shared a stratigraphic contact.

Strata 9 and 10 were excavated as a

single stratigraphic unit. These tephras

also could not be chemically linked to

known tephras in the region because of

weathering. Finally, strata 11 and 12

represent modern rootmat and forest

litter.

As mentioned previously, the profiles between the two sites are remarkably

similar and the stratigraphy described here is typical across the Susitna River floodplain

in the Trapper Creek area. However, five differences between TCO and SRO are noteworthy: 1) Stratum 9 is not as substantial at SRO as at TCO but did reach a thickness

>5cm in some parts of the site; 2) stratum 7, unoxidized loess, is rather thin on the north end of SRO where OSL samples were taken (Fig. 3.3) but this stratum does reach 15 cm in thickness in the southern units; 3) stratum 5 is less prominent at SRO than TCO; 4) stratum 3b is better developed at SRO; and 5) stratum 2 is more pronounced at SRO where it is mixed with a greater amount of fine silt than at TCO.

Four soil complexes are represented in the profile. Paleosol 1 consists of very weak A and B horizons that have developed on the basal units of strata 3, 2, and 1.

Overall, it has a very weak sub-angular blocky structure and grades from a dark grayish brown Ab horizon to an olive brown Bb horizon. The Ab horizon has moderate clay skins covering lower surfaces of clasts, while the Bb horizon has moderate clay skins covering entire clast surfaces. Paleosol 2 masks strata 5 and 4. It is characterized by strong Bbt and Bbg sub-horizons that lie immediately below the Jarvis Creek/Hayes set H tephra (stratum 6). Its features include a very weakly developed subangular blocky structure and light clay skins developed on a few pebble-sized clasts. The formation of

Paleosol 2 is coeval with a shift in vegetation demarking the emergence of a pre-boreal forest regime in the region after 9000 cal BP (Bigelow and Edwards 2001).

Paleosol 3 masks strata 8 and 7 and lies immediately below the stratum 9 tephra.

This paleosol is characterized by a brown Ab horizon that grades into a reddish brown

Bbw horizon. It is structureless but has occasional small (<2-cm diameter) mineral concretions that appear to be rich in iron oxide. The modern soil has a strong O-A horizon that has formed upon stratum 11.

Table 3.1: Stratigraphy and dating of the Trapper Creek Overlook and Susitna River Overlook sites. Trapper Creek Overlook Strata ~BS (cm) 1 Horizon 2 Description Deposition Texture/features Cal BP 3 OSL 4 12 0-3 Surface forest litter leaves, shrubs, etc. 11 3-8 O Root mat aeolian organic, roots 10 8-15 E1 Tephra leached sandy loam/ash 9 15 E2 Tephra leached sandy loam/ash, white 830-840 8 15-25 B1c Oxidized loess aeolian sandy loam/forest brown 7 25-35 B2 Unoxidized loess aeolian sandy loam 2060-2400 6 35-37 C Hayes tephra aeolian loamy sand/a few glass shards 4030 5 5* 35-40 1Ab Paleosol 2 (tephra continuous) leached sandy loam/glass devitrified 4b* 40-50 1Bb1t Paleosol 2 aeolian sandy clay loam 5870-7100 4a* 50-60 1Bb2g Paleosol 2 (tephra discontinuous) aeolian silty clay loam/gleyed 7870-8800 8120+620 3b* 60-80 2Abw Paleosol 1 aeolian sandy loam/moderate clay 3a 80-95 2Bbw Paleosol 1(very weak) aeolian loamy sand/moderate clay 21,310-22,470 2 95 2C Coarse Sand fluvial? coarse grained sand 1 >95 Ablation till ice stagnation unconsolidated clastic debris Susitna River Overlook Strata ~BS (cm) Horizon Description Deposition Texture/features Cal BP OSL 12 0-3 Surface forest litter leaves, shrubs, etc. 11 3-18 O Root mat aeolian organic, roots 10 18-25 E1 Tephra leached sandy loam/ash 9 25 E2 Tephra leached sandy loam/ash, white 8 25-35 B1c Oxidized loess aeolian sandy loam/forest brown 7 35-50 B2 Unoxidized loess aeolian sandy loam 2310-2865 6 50 C Hayes tephra aeolian loamy sand/a few glass shards 4030 5 50-55 1Ab Paleosol 2 (tephra continuous) leached sandy loam glass devitrified 4b 55-65 1Bb1t Paleosol 2 aeolian sandy clay loam 5880-6050 4a* 65-70 1Bb2g Paleosol 2 (tephra discontinuous) aeolian silty clay loam/gleyed 9140 3b* 70-85 2Abw Paleosol 1 aeolian sandy loam/moderate clay 9985-11,185 3a 85-95 2Bbw Paleosol 1 (very weak) aeolian loamy sand/moderate clay 13,990-16,280 2 95 2C Coarse sand fluvial? coarse grained sand 1 >105 Ablation till ice stagnation unconsolidated clastic debris 1 Generalized depths below surface represent top to bottom of strata. 2 Soil horizons after Graf 2006 field notes. 3 Refer to Table 3.2 and 3.3 for source data. 4 Refer to Table 3.4 for source data. 5 Hayes Set H Tephra (Schiff et al. 2008); * Artifact yielding layers.

Dating

A series of tephrochronological, radiocarbon, and OSL samples were used to date the

strata from TCO and SRO. Tephrochronological analyses were conducted at the USGS

tephrochronology laboratory at Menlo Park, California (Wan 2007) and the Alaska

Volcano Observatory in Anchorage (K. Wallace 2007 pers. comm.). OSL analysis was

undertaken by the Luminescence Dating Research Laboratory at the University of

Illinois-Chicago (Forman 2007). Charcoal samples were pretreated and AMS dated at

the University of Arizona Physics Laboratory and Beta Analytic, Incorporated.

Radiocarbon dates on charcoal provide further chronological control over the

dating of the artifact-bearing stratum at the sites. Age calibrations were calculated using

the May 2006 version of the CalPal05 calibration software (Weninger et al. 2005) and the

Intcal04 curve (Reimer et al. 2004). At both sites, charcoal fragments > 1 cm in size

were point provenienced and collected during excavation. Although very little charcoal

was recovered from SRO, the highest frequencies of charcoal at both sites occurred at the

top of stratum 4b and declined below this. TCO contained abundant dispersed and clustered charcoal horizontally across the site with some fragments reaching 3 to 4 cm in

diameter. The densest charcoal concentrations at TCO occurred in areas with some of the highest artifact densities so that they could potentially represent ephemeral hearths; however, distinguishing unequivocally between charcoal from forest fire and charcoal from human-controlled fire has not been feasible. Given the extremity of the burned features, forest fire is the most likely cause. These clustered charcoal “features” date to

5870 ± 80 and 5950 ± 40 cal BP and occurred on the surface of stratum 4b providing an accurate age for the surface of paleosol 2 at TCO. These ages are contemporaneous with

the Oshetna tephra (Child et al. 1998; Dixon and Smith 1990) suggesting a volcanic

eruption may have triggered a forest fire in the area.

Here, five new AMS 14C age estimates are reported for SRO (Table 3.2). A sample of dispersed wood charcoal from lower stratum 4a yielded a date of 8170 ± 50 BP

(9140 ± 90 cal BP). Four samples of wood charcoal provide age estimates for stratum 4b, which stratigraphically overlies the cultural component. These yielded dates of 5150 ±

40, 5160 ± 40, 5180 ± 45, and 5255 ± 40 BP. Averaged, they suggest a calibrated age of about 5955 ± 26 cal BP (5190 ± 20 BP) from the surface of stratum 4b. This date corresponds with stratum 4b at TCO. Together, these dates indicate the SRO assemblage conforms to its stratigraphic superposition beneath the Hayes tephra and is equal to or slightly pre-dates the 9140 ± 90 cal BP date obtained from stratum 4a.

Table 3.2. Radiocarbon data from the Susitna River Overlook site. Strata LAB Number BP δ13C Cal BP 1 Note Component 4b AA71144 5147+39 -22.4 5880+80 dispersed ― 4b AA71143 5162+39 -26.3 5930+40 dispersed ― 4b AA71146 5178+44 -24.9 5950+40 dispersed ― 4b AA71145 5254+39 -25.5 6050+80 dispersed ― 4a BETA208284 8170+50 -24.3 9140+90 dispersed I 1 Calibrated using CalPal05 and the Intcal04 curve at one sigma. All dates are AMS on charcoal.

AMS 14C dates (n = 10) in Table 3.3 were obtained on wood charcoal and help

define the ages of TCO’s sediments and cultural components. Five dates from stratum 4a

range from 7930 ± 40 to 7035 ± 50 BP (8800 ± 120 to 7870 ± 50 cal BP). The oldest of

these dates is from a small cluster of charcoal recovered from a pocket of ash within

stratum 4a. While this feature does not represent a hearth, it is an accurate estimate for

the first recognizable ashfall event in the area. Since this deposit caps component I at

TCO, it provides an upper bracketing age for that occupation. Charcoal samples from

stratum 4b yielded three ages. These include a date of 6200 ± 40 BP (7100 ± 70 cal BP)

on a sample of dispersed charcoal, and dates of 5190 ± 40 BP (5950 ± 40 cal BP) and

5140 ± 40 BP (5870 ± 80 cal BP) from dense charcoal concentrations on the surface of

stratum 4b. These 14C ages clearly suggest that stratum 4a at TCO dates between ca.

9150-7800 cal BP, while stratum 4b dates to about 7200-5800 cal BP. Finally, two dates from stratum 9 (890 ± 40 and 885 ± 40 BP) date the thick ash and silt layer to the late

Holocene between 800 and 900 cal BP.

Table 3.3. Radiocarbon data from the Trapper Creek Overlook site. Strata LAB Number BP δ13C Cal BP 1 Note Component 8 AA67363 884+38 -23.8 830+70 dispersed ― 8 AA67362 892+37 -23.8 840+60 dispersed ― 4b BETA208282 5140+40 -24.6 5870+80 cluster II 4b AA72199 5186+42 -23.8 5950+40 cluster II 4b BETA199720 6200+40 -23.8 7100+70 dispersed ― 4a AA67360 7035+49 -25.6 7870+50 dispersed I 4a AA67361 7068+49 -24.9 7900+50 dispersed I 4a BETA199718 7110+40 -25.0 7930+50 dispersed I 4a BETA199719 7550+40 -24.4 8370+30 dispersed I 4a BETA208283 7930+40 -23.0 8800+120 cluster I 1 Calibrated using CalPal05 and the Intcal04 curve at one sigma. All dates are AMS on charcoal.

Comparing the elemental composition of a tephra with deposits of known age is

the basic tenet of tephrochonology (Dugmore et al. 2004). Two of the three tephra layers analyzed from TCO were heavily weathered for accurate mineralogical characterization,

but the presence of biotite phenocrysts in stratum 6 match characteristics of the Hayes set

H tephra (Chapter II). The presence of these structures are the most diagnostic attribute

of the Hayes ashfall as they are not present in other known mid-Holocene tephras in the region (Riehle 1994; Riehle et al. 1990; Schiff et al. 2008:66).

Charcoal was not found below stratum 4a at either site. Consequently, sediment samples were collected for OSL dating. These numbered 16 total, five from SRO and 11 from TCO (Table 3.4). The first six OSL samples were collected at TCO in ¾-inch white

PVC tubing during the 2006 excavations. The tubes were capped on one end and

sharpened on the other. The open end was pounded horizontally into the stratigraphic

column using a rubber mallet. Sample proveniences were measured with a metric tape

relative to the site datum and ground surface. Proveniences were also documented with

an electronic theodolite, and photographs were taken of the tubes while in place. After

removal, the open ends were quickly capped and both caps secured with tape. Following

this, the tubes were placed in a brown paper bag prior to being sent to the laboratory for

analysis. Unique laboratory numbers were assigned to sediment tubes with a two digit

code at the end of the number to indicate the excitation method employed, and

backslashes separate dates on individual sediment grains within various tubes (Table 3.4).

In 2006, two tubes were placed within stratum 3a and three dates were obtained

on the lowest of these, sample UIC1865 (32,540 ± 2480 / 27,610 ± 2125 / 24,710 ± 1510

ya). Also from stratum 3a, sample UIC1935 yielded two highly aberrant dates (42,860 ±

3260 / 47,800 ± 3650 ya) from 14 cm above basal till. UIC1866 from stratum 3b was analyzed twice, each with infrared and green light excitation methods resulting in four ages in excess of 15 ka (19,530 ± 1500 / 17,685 ± 1360 cal BP and 17,450 ± 1050 /

15,580 ± 980 ya). The ages from the 2006 samples are unusually old and do not conform to stratigraphic superposition. Based on recommendations from the laboratory and general skepticism over the potential tainting of the samples by upward movement of glacial fluvial silt from strata 1 or 2, all of the 2006 samples were considered aberrant and have been dismissed. Following the 2006 season, the excavations were backfilled and, in

2007, 10 additional samples were collected from both TCO and SRO. This time, 2-inch black PVC tubing was used to collect OSL samples. At TCO, samples were collected

from the same stratigraphic column as in 2006 (Fig. 3.2), and when possible, from the

same depth below surface. In 2007, greater attention ensured samples were recovered from pure aeolian sediments with little sign of oxidation or pedogenesis.

Laboratory methods dated at least two optical ages from most of the samples;

infrared excitation was employed on components dominated by feldspar, and subsequent

tests used blue or green excitation on quartz particles. Dating methods were adjusted

when necessary between samples and included multiple aliquot regenerative dosing processes (Jain et al. 2003; Murray and Wintle 2003). Because the different methods produced ages that were statistically identical with overlapping margins of error at one standard deviation, the feldspar and quartz components of these sediments appear to be consistent geochronometers (Forman 2007).

The results from TCO supported a relatively early age for initial post-glacial loess

deposition at the site. Blue excitation was run on sample UIC2000 from the contact of

stratum 2 with 3a and yielded an age of 31,780 ± 2410 ya, similar to the 2006 samples

taken from ca. 8 cm to the left of the 2007 sample. This date is considered too old for the

base of stratum 3a and was discarded for the same reason as the 2006 dates. Additional

dates from 10 cm above stratum 2, clearly within stratum 3a, yielded more accurate

estimates for stratum 3a (22,470 ± 1720 / 21,310 ± 1630 ya). Despite the relatively old

ages for these dates, the OSL date (UIC2003) on an ash pocket in stratum 4a (8120 ± 620

ya) was consistent with 14C dates on charcoal from that horizon, suggesting the 2007

OSL calibration was accurate. Samples UIC2001 and UIC2004 from stratum 7 yielded

ages of 2400 ± 195 / < 2370 ± 160 and 2335 ± 195 / < 2060 ± 135 ya.

Table 3.4. Optically stimulated luminescence ages from TCO and SRO. Trapper Creek Overlook 1 4 Layer cm BS Lab Number Equivalent dose U (ppm) Th (ppm) K20 (%) Alpha Cosmic dose Dose rate OSL age (ya) (Grays) Efficiency2 (Grays/ka) 3 (Grays/ka) 7 30 UIC2001IR <6.86 ± 0.18 2.2 ± 0.1 5.9 ± 0.1 1.73 ± 0.02 0.08 ± 0.01 0.21 ± 0.02 3.32 ± 0.16 <2060 ± 135 7 30 UIC2001IRs 7.76 ± 0.42 2.2 ± 0.1 5.9 ± 0.1 1.73 ± 0.02 0.08 ± 0.01 0.21 ± 0.02 3.32 ± 0.16 2335 ± 195 7 32 UIC2004IR <7.43 ± 0.27 2.2 ± 0.1 5.8 ± 0.1 1.43 ± 0.01 0.10 ± 0.01 0.20 ± 0.02 3.14 ± 0.15 <2370 ± 160 7 32 UIC2004IRs 7.10 ± 0.28 2.2 ± 0.1 5.8 ± 0.1 1.43 ± 0.02 0.10 ± 0.01 0.20 ± 0.02 3.14 ± 0.15 2400 ± 195 4a 59 UIC2003IR 22.91 ± 0.20 2.8 ± 0.1 6.5 ± 0.1 1.28 ± 0.01 0.05 ± 0.01 0.20 ± 0.02 2.82 ± 0.16 8120 ± 620 3b 60 UIC1866IRs 51.21 ± 0.25 3.1 ± 0.1 5.5 ± 0.1 1.28 ± 0.01 0.07 ± 0.01 0.20 ± 0.02 1.82 ± 0.09 15,580 ± 940* 3b 60 UIC1866GRs 59.02 ± 0.34 3.1 ± 0.1 5.5 ± 0.1 1.28 ± 0.01 0.08± 0.01 0.20 ± 0.02 3.38 ± 0.13 17,450 ± 1050* 3b 60 UIC1866IR 51.21 ± 0.25 3.1 ± 0.1 5.5 ± 0.1 1.28 ± 0.01 0.07 ± 0.01 0.20 ± 0.02 2.90 ± 0.12 17,685 ± 1360* 3b 60 UIC1866GR 59.02 ± 0.34 3.1 ± 0.1 5.5 ± 0.1 1.28 ± 0.01 0.08 ± 0.01 0.20 ± 0.02 3.02 ± 0.13 19,530 ± 1500* 3a 86 UIC2002IR 70.12 ± 0.48 2.8 ± 0.1 8.6 ± 0.1 1.66 ± 0.02 0.05 ± 0.01 0.19 ± 0.02 3.62 ± 0.16 21,310 ± 1630 3a 86 UIC2002BL 77.11 ± 0.88 2.8 ± 0.1 8.6 ± 0.1 1.66 ± 0.02 0.06 ± 0.01 0.19 ± 0.02 3.43 ± 0.16 22,470 ± 1720 3a 86 UIC1935BL 142.38 ± 0.57 2.8 ± 0.1 6.7 ± 0.1 1.80 ± 0.01 0.06 ± 0.01 0.19 ± 0.02 3.67 ± 0.14 42,860 ± 3260* 3a 86 UIC1935IR 166.86 ± 1.91 2.8 ± 0.1 6.7 ± 0.1 1.80 ± 0.01 0.08 ± 0.01 0.19 ± 0.02 3.55 ± 0.14 47,800 ± 3650* 3a 95 UIC1865IRs 90.28+1.02 3.1 ± 0.1 6.8 ± 0.1 1.72 ± 0.01 0.05 ± 0.01 0.19 ± 0.02 3.80 ± 0.14 24,710 ± 1510* 3a 95 UIC1865IR 90.28 ± 1.02 3.1 ± 0.1 6.8 ± 0.1 1.72 ± 0.01 0.05 ± 0.01 0.19 ± 0.02 3.27 ± 0.13 27,610 ± 2125* 3a 95 UIC1865GR 109.32 ± 0.33 3.1 ± 0.1 6.8 ± 0.1 1.72 ± 0.01 0.06 ± 0.01 0.19 ± 0.02 3.36 ± 0.14 32,540 ± 2480* 3a 95 UIC2000BL 114.06 ± 0.75 2.8 ± 0.1 7.6 ± 0.1 1.70 ± 0.02 0.09 ± 0.01 0.19 ± 0.02 3.59 ± 0.16 31,780 ± 2410 Susitna River Overlook 7 40 UIC1991IRs 6.94 ± 0.40 2.0 ± 0.1 5.4 ± 0.1 1.58 ± 0.02 0.07 ± 0.01 0.21 ± 0.02 3.00 ± 0.15 2310 ± 185 7 40 UIC1991BL <8.95 ± 0.27 2.0 ± 0.1 5.4 ± 0.1 1.58 ± 0.02 0.11 ± 0.01 0.21 ± 0.02 3.22 ± 0.15 <2780 ± 190 7 40 UIC1991IR <8.52 ± 0.15 2.0 ± 0.1 5.4 ± 0.1 1.58 ± 0.02 0.07 ± 0.01 0.21 ± 0.02 3.00 ± 0.15 <2865 ± 190 3b 70 UIC1992IR 20.66 ± 0.13 1.9 ± 0.1 4.9 ± 0.1 0.97 ± 0.01 0.05 ± 0.01 0.20 ± 0.02 2.07 ± 0.10 9985 ± 765 3b 70 UIC1992BL 23.35 ± 0.20 1.9 ± 0.1 4.9 ± 0.1 0.97 ± 0.01 0.07 ± 0.01 0.20 ± 0.02 2.19 ± 0.10 10,675 ± 815 3b 70 UIC1994BL 28.55 ± 0.20 2.0 ± 0.1 4.9 ± 0.1 0.93 ± 0.01 0.13 ± 0.01 0.20 ± 0.02 2.55 ± 0.12 11,185 ± 865 3a 90 UIC1990IR 28.18 ± 0.12 1.6 ± 0.1 3.8 ± 0.1 1.14 ± 0.01 0.05 ± 0.01 0.19 ± 0.02 2.01 ± 0.10 13,990 ± 1070 3a 90 UIC1990BL 34.01 ± 0.18 1.6 ± 0.1 3.8 ± 0.1 1.14 ± 0.01 0.09 ± 0.01 0.19 ± 0.02 2.21 ± 0.10 15,410 ± 1175 3a 90 UIC1993IR 35.06 ± 0.19 1.7 ± 0.1 4.3 ± 0.1 1.15 ± 0.01 0.06 ± 0.01 0.19 ± 0.02 2.15 ± 0.10 16,280 ± 1240 Note: Analyses preformed by Luminescence Dating Research Laboratory, Dept. of Earth and Environmental Sciences, University of Illinois, Chicago (Forman 2007). *Samples collected in 2006 are aberrant. 1 Multiple aliquot regenerative dose methods were used for determining equivalent dose (Jain et al. 2003; Murray and Wintle 2003). Sample run under infrared (IR), green (GR), or blue (BL) excitation. 2 Alpha particle efficiency after Aitken and Bowman (1975). 3Cosmic dose rate calculated from Prescott and Hutton (1993). 4Standard deviations are at one sigma and numbers of years are from AD 2000 (ya). For all samples, moisture content was assumed to be 15 ± 5% with the exception of samples UIC1991, UIC2001 and UIC2004 which had water contents of 10 ± 3% (Forman 2007).

OSL ages from SRO support the 14C and tephra chronologies from TCO. Five

samples were extracted from the loess deposits of strata 3a, 3b, and 7 at SRO. The results

conformed to their superposition in the stratigraphic profile, and the two lowest samples

(UIC1990 and UIC1993) from within stratum 3a yielded ages of 16,280 ± 1240, and

15,410 ± 1175 / 13,990 ± 1070 ya, respectively. Two samples (UIC1992 and UIC1994)

from stratum 3b, from below the archeological component, yielded ages of 11,185 ± 865

and 10,675 ± 815 / 9985 ± 765 ya, respectively. Stratum 7 yielded three OSL ages of <

2865 ± 190, < 2780 ± 190, and 2310 ± 185 ya (UIC1991). The OSL dates from stratum

3a near its contact with 3b suggest that loess deposition on the knoll at SRO probably began after the LGM and prior to the Elmendorf stade.

Based on these results, three interpretations for the age of the sediments at TCO

can be made. First, and perhaps least confidently, the OSL ages from strata 3a suggest

that the lower loess deposits at TCO accumulated immediately after the LGM. Although

this finding may be controversial, it should not be immediately discounted. First, it

should be emphasized that these ages do not date the cultural components, but are instead

on post-glacial sediments. Second, the 8120 OSL age from stratum 4a conforms well

with AMS 14C ages on charcoal from this layer (9000-7800 cal BP). Third, stratum 7

dated to less than 2400 cal BP, an age estimate that conforms with the known age of the

Hayes tephra (ca. 4000 cal BP) in lower-lying stratum 6 and 14C dates on upper-lying tephras in stratum 9 (ca. 800-900 cal BP).

The 14C, OSL, and tephrochronological results described above offer excellent chronological control over the geological deposits and cultural components preserved at

SRO and TCO. With the exception of the 2006 OSL ages, all of the ages correlate by

depth below the surface suggesting cryoturbation and other post-depositional deformation processes have been limited throughout most of the sediments. In the lowest post-glacial loess deposits, the OSL data suggest deposition may have begun earlier in the late

Pleistocene than anticipated. At the onset of this project, deglaciation in the Trapper

Creek area was assumed to have occurred in the late glacial based on the glacial chronology of Cook Inlet, Kenai Peninsula (Reger et al. 1995; et al. 1996; et al. 2007), and the ages of the Riley Creek readvance in Broad Pass (Briner et al. 2008; Dortch

2006). Because these regions bracket the study region to the north and south, it was assumed the middle Susitna River region was glaciated as late as 14 ka; however, the

OSL dates from the Trapper Creek sites suggested otherwise.

The middle of stratum 3a at SRO dates as early as ca. 16.28 ka and lower sections of this stratum at TCO where the horizon is thickest appears to date as early as ca. 22 ka.

It can be tentatively concluded that the tri-river area of southcentral Alaska became ice- free early in the late glacial, immediately after the LGM, and was not glaciated during the

Elmendorf readvance. The ages of strata 3b and 4 at both sites are pertinent to the cultural chronology of the site, since artifacts were only encountered in these layers.

Based on data presented here, the loess deposits in stratum 3b accumulated between 11.3 and ca. 9.2 ka. Unfortunately, precision is lacking on the basal age of Paleosol 1 but the transition to Paleosol 2, stratum 4a, is coeval with a shift in the vegetation during the birch rise ca. 9100 cal BP in central Alaska (Chapter II). Stratum 4b dates between 7.2 and 5.9 ka. Dense charcoal clusters on the surface of Paleosol 2 (stratum 4b) are coincident with the arrival of black spruce to the area and a consequential increase in

incidence and intensity of natural forest fires across southcentral Alaska (Lynch et al.

2002) as well as the Oshetna tephra marker.

Given the chronologies and artifact proveniences discussed above, it can be

concluded that the age of the SRO occupation is minimally 9200-9100 cal BP because

artifacts were recovered in situ on the surface of stratum 3b. The age of TCO-I is slightly

older than SRO based on the slightly deeper position of in situ artifacts recovered from 5

cm below the surface of stratum 3b at TCO while the component at SRO was deposited

directly on the surface of stratum 3b. The TCO-I occupation persisted until ca. 7800 cal

BP and the component II assemblage dates between 7100-5800 cal BP.

No artifacts have been recovered at either site above stratum 4b, since the deposition of the stratum 5 tephra. Two potential scenarios could explain this trend: 1) foraging societies were exterminated or vacated the region because of increased forest fires and volcanism in the middle Holocene (Lynch et al. 2002; Schiff et al. 2008); or 2) subsistence and landuse strategies shifted from large animal hunting to salmon harvesting so that hunting overlooks were no long occupied. During the 2004 survey, many

Athapaskan house and cache pit sites were discovered and none were older than 800 cal

BP, but excavations of these sites have not been undertaken, so no sites dated from 5800 to 800 cal BP are known from the area. Further research into the middle Holocene occupation of southcentral Alaska will ultimately determine which if these scenarios are more accurate.

Artifact Assemblages

The SRO assemblage was deposited during a single occupation event and consists of 404 pieces of flaked debris and 36 tools (sum = 440 lithic pieces). Artifacts in both TCO

components were deposited by small groups via a series of intermittent palimpsest

events. Stratigraphic separation allowed categorizing the artifacts into two separate

components with a hiatus between 7800 and 7100 cal BP. The earliest of these, TCO-I,

consists of 267 lithic pieces, including 252 flaked debris and 15 tools. TCO-II contains

340 lithic pieces, 308 flaked debris and 32 tools.

Methods

My analysis examined attributes of tools and flaked debris at the TCO and SRO sites.

Tools are defined (or debitage) as items that were used and are distinguished from flaked

debris, defined as detritus and discarded without use. Attributes analyzed include raw

material type, degree of dorsal cortex, platform preparation, metrics, and weight for all

artifacts. Tools were further analyzed according to edge angle, condition, and retouch

attributes that included degree of invasiveness, retouch form, location, and number of

retouched margins. Tool assemblages were further distinguished by formal and informal

varieties. Formal tools were those that were curated as opposed to informal or expedient tools which were not. See Appendix for frequencies of size classes, debitage, and tool types from all components.

Raw materials were analyzed based on visual inspection of material type, color, and texture. Representative samples were taken to the Geosciences Department at

University Nevada, Reno for further classification. The degree of cortex was measured by estimating the percentage of the dorsal surface area with cortex and classifying these in five categories ranging from 0 to greater than 90 percent. Tool size was categorized according to 3 cm artifact size classes ranging from less than 1 cm to 29-31 cm using a concentric circle template. Platform preparation occurred into five forms including

simple preparation defined as smooth or straight, complex with multiple surface facets,

crushed, cortical, and unidentifiable or absent platforms. Debitage was classified

according to 18 separate types or classes and included documentation of fragmented and complete flakes. Cores and microblades were included in the debitage analysis since cores are related to artifact production and many microblades and microblade segments are discarded without use after being detached.

Tool metrics included length, width, and thickness measured with digital caliper.

Weights to the nearest 0.1-g up to 200-g were taken with a precision digital scale; and

weights over 200-g were taken using a balance scale with calibration weights. Edge

angles were documented to the nearest degree with a goniometer. Retouch invasiveness,

measured with digital calipers, documented the greatest extent of retouch from the lateral

margin to the center of the artifact. Number of retouched margins is a score based on the

number of tool margins with retouch. Retouch location was classified according to where

retouch occurred on the tool, and retouch form included stepped, scalar, marginal

grinding, use wear only, and other.

Susitna River Overlook

Artifacts at SRO were recovered in strata 3b and 4a, but because all of the artifacts >700

g were found in situ on the surface of stratum 3b, it is assumed this was the original

occupation surface and the assemblage was later covered by stratum 4a. Artifacts

recovered from 4a were few in number, within 5 cm of stratum 3b, and tended to be

vertically aligned suggesting upward transport from below. The potential for palimpsest occupations has been eliminated based on refitting fragments between artifacts, including

segments of a finished biface recovered from between the two strata (Fig. 3.4).

Figure 3.4. Distribution of point provenienced artifacts and refits at SRO. Photo depicts (a) in situ cluster of chopping implements and flake cores; (b) disturbed ground from the initial 2004 shovel probe that led the site discovery; (c) dispersed charcoal yielding a date of 9140 ± 90 cal BP on the base of stratum 4a; (d) exposed paleosurface of stratum 3b; and (e) two fragments of a fractured finished biface and side scraper vertically aligned and only a few cm above stratum 3b.

Table 3.5. Raw material types at SRO. Debitage Tools Total Raw Material n % n % n % Basalt 295 73 21 58.3 316 71.8 Chalcedony — — — — — — Chert 18 4.5 3 8.3 21 4.8 Obsidian — — — — — — Sandstone 65 16.1 7 19.4 72 16.4 Schist — — — — — — Siltstone 26 6.4 4 11.1 30 6.8 Slate — — — — — — Quartzite — — 1 2.8 1 0.2 Total 404 100 36 100 440 100

Raw material at SRO is primarily basalt (71.8%), followed by sandstone (16.4%),

siltstone (6.8%), chert (4.8%), and quartzite (0.2%). Among tools, basalt (58.3%) is

again dominant followed by sandstone (19.4%), siltstone (11.1%), chert (8.3%), and

quartzite (2.8%). Flaked detritus and cores at SRO were overwhelming basalt (73%),

followed by sandstone (16.1%), siltstone (6.4%), and chert (4.5%) (Table 3.5).

The preponderance of tools at SRO are informal (91.7% of tools), primarily flake tools (27.8%) produced during the construction of heavy bifacial (13.9%) and unifacial

(11.1%) chopping implements. Other informal tools include cobble tools (16.7%), retouched flakes (11.1%), tci-thos (5.6%), and a utilized blade (2.8%). Retouched flakes tend to be unifacially worked with marginal retouch that did not significantly alter the flake. Formal tools (8.3%) that were fractured and abandoned at the site include unhafted biface fragments (5.6%) made on siltstone and two side scrapers (5.6%) on a fine-grained volcanic material, classified as basalt but similar in texture to obsidian. Refitted biface fragments were combined and analyzed as a single tool (Fig. 3.5). The two side scrapers include a double straight sided scraper constructed on a large blade-like flake with unifacial retouch on one end and both lateral margins with an acute

(40°) edge angle. The second scraper was also unifacial with retouch on one end and along one lateral margin with a relatively steep edge angle (70°). Figure 3.5. Refitting fragments of a biface from SRO. (sketch by E. Pellegrini)

Figure 3.6. Heavy choppers from SRO including (a) a bifacial and (b) unifacial chopper with edge damage. (sketch by E. Pellegrini)

The majority of tools at SRO are complete or unbroken (72.2%). Tools were often constructed on flakes (58.3%) and cobbles (30.6%) and there is a relatively high percentage

(30.6%) of tools with more than 50% dorsal cortex. Of these, 22.2% have cortex on 90% of the dorsal surface. In contrast, the bulk of flaked debris (89.9%) lacks dorsal cortex.

Tools were frequently retouched on just one edge (55.6%), while 22% have two worked edges, 16.7% have three, and 5.6% have four worked edges.

Retouch is primarily in the form of use wear only (55.6%), followed by stepped

(25%), scalar (16.7%), and pitted (2.8%) forms. Significantly, 50% of the tools are larger than 11 cm in size and a relatively small number are less than 3 cm (8.3%). Flaked debris at SRO is also relatively large with 50.2% greater than 3 cm. A third of the tools do not have platforms but when present most are cortical (33.3%), followed by simple (25%), and only five are complex (8.3%); none of the tools has evidence of a crushed platform. On flaked debris, platform preparation is primarily unidentifiable (45.3%) given the heavily fragmented and weathered nature of the SRO assemblage. It is assumed most of these had cortical platforms. In cases where platforms are present on flaked debris, 7.4% are cortical,

28.7% simple, 18.3% complex, and 0.2% crushed.

Flaked debris consists of complete (31.4%), fragmented flakes (29.2%), and one blade-like-flake (0.2%) produced on site primarily from silt and sandstone cores that were then used as heavy choppers. Evidence for initial core reduction includes angular shatter

(6.2%), as well as primary (2.7%) and secondary cortical spalls (5.2%), and a cobble fragment (0.2%). Biface thinning flakes (15.3%), retouch chips (7.4%), and chip fragments

(2%) indicate that secondary reduction activities (i.e., tool resharpening) also occurred. No evidence of microblade production or composite tool repair (i.e., microblade cores, core

rejuvenation flakes, burins, burin spalls, or microblade segments) was found among the flaked debris at SRO.

Trapper Creek Overlook, Component I

The TCO-I assemblage suggests the initial occupants engaged in discrete refurbishing activities with tool types including microblades and microblade cores, a burin, and a variety of small flake tools. Several clusters in TCO-I fit Binford’s (1983:152) description for individual tool maintenance tasks in which small distributions of items are dropped and tossed around the worker in an arc formation (Krasinski and Yesner 2008). Doug

Anderson (1988) noted at Onion Portage that these clusters generally fell within 2 m diameter areas for single activities. Although no formal spatial analysis has been undertaken, discrete artifact clusters in TCO-I are discernable (Fig. 3.7).

At TCO-I, there are a variety of chert types (50.9%), followed by basalt (30%), and two types of obsidian (2.2%) sourced by X-ray fluorescence to the Batza Téna obsidian quarry (R. J. Speakman 2006 pers. comm.). Schist (10.5%) was utilized in relatively high frequencies in TCO-I, probably because the material is plentiful in ablation till at the site.

Slate (1.5%) was also used in small amounts and occurs naturally at the base of the Trapper

Creek bog near the site, and small amounts of siltstone (2.2%), sandstone (1.5%), and chalcedony (1.1%) occur in TCO-I. Chert (46.7%) dominated the tool assemblage followed by basalt (33.3%), obsidian (13.3%), and chalcedony (6.7%). Flaked detritus and cores were primarily chert (51.2%), followed in frequency by basalt (29.8%), siltstone

(2.4%), obsidian (1.6%), sandstone (1.6%), slate (1.6%), and chalcedony (0.8%) (Table

3.6).

Table 3.6. Raw material types at TCO-I. Debitage Tools Total Raw Material n % n % n % Basalt 75 29.8 5 33.3 80 30 Chalcedony 2 0.8 1 6.7 3 1.1 Chert 129 51.2 7 46.7 136 50.9 Obsidian 4 1.6 2 13.3 6 2.2 Sandstone 4 1.6 — — 4 1.5 Schist 28 11.1 — — 28 10.5 Siltstone 6 2.4 — — 6 2.2 Slate 4 1.6 — — 4 1.5 Quartzite — — — — — — Total 252 100 15 100 267 100

TCO Component I Debitage Tool

meters Figure 3.7. Point provenienced artifacts at TCO-I.

Formal tools in the TCO-I assemblage include the tip of a crystal clear obsidian point (6.7%), an unhafted thinned biface (6.7%) on basalt, an end scraper (6.7%) on tan chert with a possible burin blow on one edge, and two fragmented side scrapers (13.3%), one on brown chert and one on chalcedony. Informal tools include six flake tools (33.3%), a utilized blade (6.7%), notched flake (6.7%), tci-tho (6.7%), hammerstone (6.7%), and a burinated flake (6.7%). The majority of tools at TCO-I are complete and constructed on flakes (73.3%). No tools have more than 50% dorsal cortex and most (73.3%) had no cortex at all. A similar pattern was found among the flaked debris where there is a high percentage (92.9%) lacking cortex. Tool retouch generally occurred on just one edge

(53.3%), one third (33.3%) have two worked edges, and 13.3% have three. Retouch is primarily in the form of use wear only (53.3%), followed by scalar (40%), stepped (25%), and marginal grinding (6.7%) forms. Only one tool is larger than 11 cm in size and three are less than 3 cm. Flaked debris is also relatively small with 84.5% less than 3 cm. Forty percent of the tools lack a preserved platform but when present platforms are either complex (46.7%) or simple (13.3%); none are crushed or cortical. Platform preparation on flaked debris is more variable and includes unidentifiable (37.3%), simple (31.3%), complex (26.2%), cortical (3.6%), and crushed (1.2%) platforms.

Semi-diagnostic microblade cores from TCO-I are characterized as an end-style microblade core fashioned from a recycled end-scraper and a conical shaped core (Fig.

3.8). The striking platform on the end core was prepared with light retouch and grinding the back lateral margins of a flat, non-beveled platform. No core tablet was removed. One partial and three full microblade facets occurred on the core face. The other microblade core was conical in style with six blade facets encircling the entire core except on one

cortical face. Platform preparation on this specimen was simple light retouching with no evidence of tabular rejuvenation.

Figure 3.8. Lithic artifacts from TCO-I: (a) end-style microblade core; (b) conical microblade core; (c) burin; and (d) microblades. These artifacts were recovered from at least 5 cm below the surface of strata 3b and were found in undisturbed contexts. (sketch by E. Pellegrini)

Flaked debris was highly fragmented with complete (9.9%) and fragmented (27%) flakes, retouch chips (16.7%) and chip fragments (2.0%), bifacial thinning flakes (20.6%), microblades (4.0%), burin spalls (1.2%), blade-like-flake (0.8%), and bipolar flakes

(0.4%). Both primary and secondary reduction is represented in the core and flaked debris at TCO-I. Cores and core fragments include two microblade cores (0.8%) and two microblade core face rejuvenation flakes (0.8%). Lithic reduction debris included angular shatter (9.9%), primary (2.4%), and secondary (1.6%) cortical spalls, two split cobbles and a worked chert cobble (1.2%). Flaked debris occurs on chert (49.2%), basalt (31.3%), and

schist (11.7%), followed by lesser numbers of obsidian (1.7%), chalcedony (0.4%), sandstone (1.7%), and slate (1.7%).

Trapper Creek Overlook, Component II

The middle Holocene occupation of TCO-II is more diffuse (Fig. 3.9), perhaps the result of more frequent visits to the site. Diffuse patterns were described by Binford (1983) as the result of extensive, rather than discrete, activities. During butchering and hide processing, the worker(s) moves around the task area, discarding lithics and shuffling debris resulting in wider and more strewn patterns. In these circumstances, discarded material spans larger areas and is less densely clustered than debris generated during discrete activities. Further indications of extensive activity areas are “blank areas” within scatters, a potential consequence of the event focus. For example, areas void of artifacts within diffuse lithic scatters may occur where hides or carcasses were on the ground (Binford 1983:165-172).

Similar patterns associated with hide processing activities may have occurred at Broken

Mammoth in interior Alaska (Krasinski and Yesner 2008).

TCO-II is the most diverse of the three assemblages with basalt (39.7%), chert

(28.2%), two varieties of obsidian (14.1%), schist (6.5%), siltstone (5.9%), sandstone

(2.1%), slate (1.8%), and chalcedony (1.8%). Many tools were constructed on basalt

(31.3%) and chert (28.1%) followed by siltstone (18.8%), obsidian (9.4%), sandstone

(6.3%), slate (3.1%), and chalcedony (3.1%). Flaked debris consists of basalt (40.6%), chert (28.2%), obsidian (14.6%), schist (7.1%), siltstone (4.5%), sandstone (1.6%), slate

(1.6%), and chalcedony (1.6%) (Table 3.7).

Table 3.7. Raw material types at TCO-II. Debitage Tools Total Raw Material n % n % n % Basalt 125 40.6 10 31.3 135 39.7 Chalcedony 5 1.6 1 3.1 6 1.8 Chert 87 28.2 9 28.1 96 28.2 Obsidian 45 14.6 3 9.4 48 14.1 Sandstone 5 1.6 2 6.3 7 2.1 Schist 22 7.1 — — 22 6.5 Siltstone 14 4.5 6 18.8 20 5.9 Slate 5 1.6 1 3.1 6 1.8 Quartzite — — — — — — Total 308 100 32 100 340 100

TCO Component II Debitage Tool

meters Figure 3.9. Point provenienced artifacts at TCO-II.

Formal tools at TCO-II include three unhafted bifaces (9.4%), two on basalt and one on tan chert, an end scraper (3.1%) on basalt, and four side scrapers (12.5%), three on chert and one on slate. Informal tools include retouched flakes (25%), flake tools (12.5%), tci-thos (18.8%), and cobble tools (18.8%). The majority of tools in TCO-II are complete

(59.4%) as opposed to fragmented (40.6%), and more than half were constructed on flakes

(53.1%). Many tools lacked dorsal cortex (73.3%). Where cortex was present, six (18.8% of the tool assemblage) had more than 90% dorsal surface coverage, two (6.3%) had 10 to

50% surface coverage, and one (3.1%) had 1 to 10% dorsal cortex. Among the flaked debris, a high percentage (87.3%) lacked cortex. Only a few flakes (7.9%) had between

1% and 90% dorsal cortex and fifteen flakes (4.9%) had greater than 90% dorsal cortex.

Tool retouch was primarily on one edge (59.4%), 31.3% have two worked edges, 6.3% have three, and 3.1% of tools have retouch on four edges. Retouch is primarily scalar

(34.4%), followed by use wear (31.3%), pitting (15.6%), marginal grinding (12.5%), and stepped (6.3%) forms. There are no tools smaller than 1 cm in diameter and 9.4% are between 1 and 3 cm in size, but the majority are between 3 and 5 cm (28.1%) or 5 and 7 cm

(21.9%). Several are between 7 and 11 cm (31.3%), very few tools are larger than 11 cm in size (9.4%) and none are larger than 15 cm. Flaked debris is also relatively small with

85.7% less than 3 cm. Many tools lack a preserved platform (40.6%), but when present they are generally cortical (21.9%), followed in proportion by simple (18.8%), complex

(15.6%) or crushed (3.1%). Platform preparation on flaked debris includes simple (38%), complex (23.4%), cortical (5.8%), and crushed (2.6%) varieties; 30.2% lacked platforms.

Flaked debris consists of complete (13%) and fragmented flakes (20.1%), retouch chips (21.4%) and chip fragments (5.4%), bifacial thinning flakes (18.2%), microblades

(4.2%), and smaller amounts of burin spalls (0.6%), blade-like-flakes (0.6%), and bipolar flakes (0.6%). Elements of primary and secondary reduction include cobble fragments

(1.3%), multidirectional flake cores (1%), split cobbles (0.6%), worked chert cobbles

(0.3%), angular shatter (4.2%), primary (3.6%), and secondary (4.5%) cortical spalls.

Discussion

Classifying the Trapper Creek assemblages into known Alaskan archeological techno- complexes is difficult because of ongoing disagreement and uncertainty pertaining to assemblage variability in the early prehistory of central Alaska (Goebel and Buvit 2009).

Tool types common between the assemblages include cobble tools, tci-thos, side scrapers, unhafted bifaces, and utilized flakes (Fig. 3.10). The tci-thos are disk shaped split cobble scrapers or knives with high amounts of dorsal cortex manufactured via an expedient bread-loafing technique that produces large primary and secondary cortical spalls from river cobbles (Fig. 3.11). Tci-thos recovered at TCO and SRO have slight varnish and edge wear consistent with hide scraping or flesh cutting.

Tci-thos are common throughout interior Alaska where they have been found alongside microblades in the late Holocene component at Swan Point (Holmes et al. 1996), and at Panguingue Creek component II with scrapers, burins, choppers, and hammer and anvil stones (Goebel and Bigelow 1996), for example. These items are also common in the

Denali complex assemblages at the Tangle Lakes (West 1996 a, b, c), and at the Ravine site where they occur with unusually large flaked debris (> 18 cm), macro and microblade technology, end and side scrapers as well as lenticular bifaces (Robinson et al. 1996). A similar tool type, known as skreblos, are often reported alongside heavy chopping tools and microblades in Siberian sites (Derevianko 1996:286; Dikov 1996:248; Goebel 1999, 2002;

Larichev et al. 1992; Medvedev and Merbs 1964). Thus, this array of heavy chopping, smashing, and scraping tools including the tci-thos is frequent among early and late microblade industries in both eastern and western Beringia and should be regarded as a sub-complex related to specific butchering and game processing activities.

Chopper (uniface) Chopper (biface) Notch Cobble Tool Tci-thos Side Scraper End scraper Burin Utilized Blade Retouched flake Flake tool Biface unhafted Bifacial Point

0% 5% 10% 15% 20% 25% 30% 35%

SRO TCO-I TCO-II

Core rejuvenation flake Mulidirectional flake core Microblade core Core fragment Worked chert cobble Split cobble Cobble fragment Angular shatter Burin spall Bifacial thinning flake Retouch chip Retouch chip frag. Sec. Cortical spall Prim. Cortical spall Microblade Blade-like-flake Bipolar flake Complete flake Flake fragment

0% 5% 10% 15% 20% 25% 30% 35%

SRO TCO-I TCO-II

Figure 3.10. Frequencies of tool types (top) and flaked debris (bottom) from TCO and SRO.

Figure 3.11. Example of refitting split cobble tci-tho implements from TCO-II. Image depicts dorsal and ventral surfaces. Note the “bread-loafing” reduction technique.

TCO-I has the greatest tool diversity in different tool types. It is also the only one of these assemblages with evidence of a bifacial projectile point industry consisting of the tip fragment of an ultra-thin and delicately made point on transparent and slightly smoky

Batza Téna type K obsidian (R. J. Speakman 2006 pers. comm.). It also has a higher percentage of flake tools despite having the lowest frequency of complete flakes. Among elements of microblade technology, the end-style microblade core from TCO-I is similar to a style described by Powers (1973) from Paleolithic Siberia and occasionally reported in

Denali complex sites of interior Alaska (West 1981). Although TCO-II does have microblade fragments and burin spalls, it lacks diagnostic cores. A preponderance of retouch and biface thinning flakes in the TCO assemblages indicates tool maintenance and repair were important activities in both components. In addition to refurbishing, hide processing was a significant task during TCO-II times based on the diffuse nature of the assemblage and a high prevalence of tci-thos, scrapers, and retouched flakes.

Although faunal remains were not preserved at SRO, the in situ position and refit analysis of flaked tools and heavy chopping implements leads to the interpretation that the occupation was a short-lived, single occupation event probably the result of large mammal butchering and marrow extraction. This hypothesis is further supported by the overall size and position of the large flake tools, various scrapers and heavy choppers, some still resting on anvils and most with battered edge damage. Tools found at SRO are consistent with other known special butchering assemblages from the central Alaska Range including an activity area at Dry Creek component II (Powers et al. 1983), Carlo Creek (Bowers 1978;

Bowers and Mason 1992), and Panguingue Creek (Goebel and Bigelow 1996), each located within 215 km north of Trapper Creek. It also bears similarities to the tentatively proposed

Amphitheater Mountain complex from Tangle Lakes (West 1974). Although the purpose of those assemblages has been disputed (Bever 2001b:139, Mobley 1982; Yesner and

Pearson 2002:139), they most likely represent specialized game processing assemblages.

Flake size, degree of cortex, and raw material types are frequently featured in lithic analyses to explain landuse, mobility, and raw material procurement strategies in prehistoric contexts (Andrefsky 1994, 1998; Blades 1999; Clarkson 2002; Close 2000;

Feder 1980; Kelly 1992; Kuhn 1995; Newman 1994; Odell 1996, 2004). Greater precision has come from obsidian sourcing projects which use neutron activation and X-ray fluorescence to precisely link unique chemical compositions in artifacts to their point of origin if the quarry source is known (Craig et al. 2007; Glasscock et al. 2007). Integral to these types of analysis is an assessment of the utilization of local and non-local raw materials.

Systematic surveys for locally available raw materials in the vicinity of Trapper

Creek were undertaken by S. Coffman (2006) and students of the UNR archeological fieldschools. Random identification of river cobbles in gravel bars along the major rivers was undertaken both upstream and downstream of their confluence east of TCO. Smaller tributaries in the vicinity of the sites were also studied. The students discovered that all of the raw material types found in the SRO and TCO assemblage, except the obsidians, are locally available within a few kilometers of the sites. It is assumed these resources were known and used by prehistoric foragers in the region.

Basalt is the primary lithic type in SRO and TCO-II, followed by chert. The reverse pattern occurs in TCO-I where chert is more frequent (Figs. 3.12 and 3.13).

Chemical analysis of the two varieties of obsidian found at the Trapper Creek site matched

Batza Téna sources (R. J. Speakman 2006 pers. comm.). Batza Téna, located 420 km northwest of Trapper Creek along the Koyukon River, was the most commonly used obsidian source in Alaska with samples distributed across the state and dated to as early as

the late Pleistocene (Clark 1995; Clark and Clark 1991; Cook 1995). In TCO-I, much of the mahogany obsidian (type unknown) from Batza Téna refit into a single shattered flake core. The second variety was smoky grey to completely transparent in color (type K) and consisted only of retouch chips and the projectile point tip.

Quartzite Slate Siltstone Schist Sandstone Obsidian Chert Chalcedony Basalt

0% 10% 20% 30% 40% 50% 60%

SRO TCO-I TCO-II

Figure 3.12. Frequency of raw materials among tools at TCO and SRO.

Quartzite Slate Siltstone Schist Sandstone Obsidian Chert Chalcedony Basalt

0% 10% 20% 30% 40% 50% 60% 70% 80%

SRO TCO-I TCO-II

Figure 3.13. Frequency of raw materials among flaked debris at TCO and SRO.

A comparison of the frequency of lithic types among tools relative to flaked debris

(depicted as a ratio) is a relative indicator of the number of tools worked but not deposited

in a site, thus a proxy for tool curation. At SRO the tool-to-debitage ratio of raw material is

5:4 (five different raw materials represented in the tool assemblage and four among the flaked debris). This ratio in TCO-I is 4:8 and in TCO-II it is 7:8. TCO-I stands apart with twice as many raw materials represented in the flaked debris than among the tools, an indication that significantly more tools were worked or constructed at the site and transported away than were left behind. This pattern suggests a highly curated lithic technology because of increased raw material conservation that may represent several factors: 1) high mobility; 2) limited knowledge of local raw material sources; 3) a lack of toolstone on the landscape; or 4) a seasonal shortage of toolstone. Because raw material is plentiful and easily found across the landscape, options 2 and 3 are not tenable. Although option 1 is widely applicable among Paleoindian sites in southern North America where raw material sources are discrete and widely dispersed (Kelly 1992; Kelly and Todd 1988), it is less applicable in northern latitudes where the distribution of raw material is different.

Thus, option 4 is the most parsimonious explanation for the conservation behavior inherent in the TCO-I assemblage.

The frequency of raw material types among complete and fragmented microblades from TCO indicates a minimum number of separate microblade cores represented in each component. All of the microblades were on chert except two, one being on obsidian and the other on tan chalcedony. These were categorized by color to further distinguish individual chert specimens under the assumption that different colors and textures indicated unique parent materials or cores (Table 3.8).

Table 3.8. Microblade raw material at TCO. TCO CI TCO CII Totals Material/color n % n % n % Chert/tan 5 50.0 4 30.8 9 39.1 Chert/red 1 10.0 2 15.4 3 13.0 Chert/white 2 20.0 2 15.4 4 17.4 Chert/grey 1 10.0 2 15.4 3 13.0 Chert/green — — 1 7.7 1 4.3 Chert/black — — 1 7.7 1 4.3 Chalcedony/tan 1 10.0 — — 1 4.3 Obsidian/black — — 1 7.7 1 4.3 Total 10 100 13 100 23 100

Using this approach, a minimum of five separate microblade cores were represented at TCO-I and seven in component II. These figures do not necessarily reflect how many cores were flaked at the site, only the minimum number of cores involved in producing the microblades recovered there. Because many of the microblades were fractured, the majority may have been produced elsewhere and transported to the site where they were deposited through refurbishing activities. Of the microblade fragments recovered in both components at TCO, only nine tan chert fragments possessed the toolstone characteristics of a recovered core. No microblade fragments of the same orange chert toolstone as the end-style microblade core were recovered; however, one core rejuvenation flake of that material was found.

In situations where raw material sources occur in distinct locations on the landscape, flake sizes have proven useful in predicting the relative distance of sites from quarry sources (Feder 1980; Newman 1994). However, the relationship between flake size and distance from a quarry does not always reflect a direct relationship. Alaskan archeologists have been hesitant to apply this concept in northern latitudes because raw material tends to be plentiful and spread across the region in gravel bars. Recent studies

(Blades 2003:153-154) disassociate increased reduction in the form of flake size with

increased mobility or distance from raw material source primarily because there are more pressing factors influencing reduction intensity. For example, in Chapter V, I argue that time since procurement and not distance from the quarry is a more significant factor in

Alaska where raw material is generally readily available on the landscape in streams and gravel bars except in the winter months when these resources are locked beneath ice and snow. These factors are considered in the following analysis of artifact size and degree of dorsal cortex as proxies for time since procurement, since the majority of raw material recovered from SRO and TCO was locally available.

Tools at SRO tend to be larger with more dorsal cortex than in the TCO assemblages (Fig. 3.14, Fig. 3.16). Most tools in TCO-I and a third of the tools in TCO-II were less than 5 cm in size. The pattern among flaked debris is similar with most debitage less than 3 cm in the components at TCO. In contrast, at SRO flaked debris tends to be larger (Fig. 3.17). The relatively small amount of cortex and small size classes among the tools and debitage in the TCO assemblages (Fig. 3.14 and 3.15) are indications of extensive retouch, tool reworking, and raw material conservation. Although frequencies of cortex among flaked debris are similar at SRO, the opposite probably occurred given the overall size and refits among tools, it is clear toolstone was procured nearby with the majority of tools produced and discarded onsite. Production and maintenance of curated tools was not the primary activity at SRO nor was raw material conservation.

The only local sources on the landscape with cobbles of a size capable of producing the SRO choppers occur along high velocity rivers. Today, the nearest available gravel bar with boulders of 30 cm in diameter or larger occur several km downstream along the

Susitna River, but in prehistory cobbles of that size would have been available in the fossil slough immediately west of the site (Fig. 3.1).

80.0%

60.0%

40.0%

20.0%

0.0% 0% 1-10% 10-50% 50-90% >90%

SRO TCO-I TCO-II

Figure 3.14. Degree of dorsal cortex on tools at TCO and SRO.

100.0% 80.0% 60.0% 40.0% 20.0% 0.0% 0% 1-10% 10-50% 50-90% >90%

SRO TCO-I TCO-II

Figure 3.15. Degree of dorsal cortex on debitage at TCO and SRO.

40.0% 30.0% 20.0% 10.0% 0.0% 3 5 - 7 - 9 < 1 < - 1 - 3 5 11 7 13 - 15 - 17 - 9 21 - 23 - 25 - 27 11 - 13 29 - 15 31 - 19 - 21 23 25 27 29

SRO TCO-I TCO-II

Figure 3.16. Tool size classes (cm) at TCO and SRO.

60.0%

40.0%

20.0%

0.0%

< 1 1-3 3-5 5-7 7-9 9-11 11-13 13-15

SRO TCO-I TCO-II

Figure 3.17. Debitage size classes (cm) at TCO and SRO.

Platform preparation on tools is predominately cortical and simple at SRO.

Platforms tend to be complex on tools from TCO-I, and at TCO-II there are relatively equal proportions of cortical, simple, and complex platforms on tools (Fig. 3.18). Among flaked debris, the platform preparation is uniform between assemblages with simple and complex styles dominating (Fig. 3.19). Higher frequencies of complex platforms on tools is a further indication that flaking of toolstone was done more carefully at TCO-I than at SRO where cortical platforms were more common.

50.0% 40.0% 30.0% 20.0% 10.0% 0.0%

SRO TCO-I TCO-II

Figure 3.18. Tool platform preparation at TCO and SRO.

50.0% 40.0% 30.0% 20.0% 10.0% 0.0%

SRO TCO-I TCO-II

Figure 3.19. Platform preparation on flaked debris at TCO and SRO.

The attributes described distinguish two types of sites with different approaches to raw material use in the Susitna River lowlands. First is the relatively conservative use of lithic resources at the two TCO components where refurbishing activities predominate.

TCO-I is a small assemblage with tools and flaked debris suggesting a few separate occupation events related to hunting overlooks and composite tool technology, while TCO-

II, which is spatially more diffuse and has a wider variety of tools and raw materials, is indicative of hide processing activities. The second type of occupation is the game butchering assemblage from SRO. At SRO, a high number of refits and low raw material diversity imply a single occupation event where tools were expediently constructed and discarded in a tightly clustered pattern. Because the toolstone source was probably nearby, conservation of raw material was not a priority. This does not appear to be the case in the

TCO-I assemblage where high tool curation, ultra-small tools including microblade technology, and virtually no dorsal cortex on flaked tools suggest a degree of lithic stress that influenced how refurbishing occurred despite the local availability of most lithic types found there.

Because prehistoric foragers were probably aware of the raw materials available in the area, seasonal shortages during winter months could explain the nature of the TCO assemblages. None of these occupations represents a semi-permanent base camp with established hearth features, tent or semi-subterranean features, and caches (Guthrie 1983).

Instead, they may represent spike camps or temporary, task specific occupations oriented toward resource extraction, suggesting these groups were logistically organized and oriented toward a subsistence system in which small specialized task forces set out with specific resource procurement activities in mind. Whether these resources were then collected and transported back to a more permanent base camp where they were shared with others in the community, or alternatively if resources were consumed near the procurement site before moving on, cannot be determined given the dearth of known sites in the area.

Conclusions

Large sections of southcentral Alaska have remained relatively unknown to archeology, primarily because the dense forest canopy restricts aerial and pedestrian survey across the region. In 2004, endeavors against this trend resulted in the discovery of two buried early

Holocene sites near the community of Trapper Creek. The geoarcheological investigations which followed suggest subarctic foragers with ties to the north were the first to colonize the Susitna River lowlands by following major rivers south out of prominent mountain passes. Radiocarbon, OSL, and tephrochronological data suggest these events took place in the vicinity of Trapper Creek sometime between ca. 11,000 and 9200 cal BP. Based on the evidence presented here, the first human groups to arrive in the Susitna River lowlands were small and engaged primarily in specialized hunting and game processing related

activities. This suggests a degree of logistical organization and mobility by experienced groups with specific tasks in mind, patterns similar to late Pleistocene sites in the adjacent

Nenana and Tanana Valleys to the north. Questions remain whether these events led to a sustained colonization of the region or if it was vacated in favor of Alaska’s southern coasts in the middle Holocene.

Based on the nature and distribution of the assemblages, it can be concluded that the initial occupations were small and sporadic in scale. Subsequent occupations in the middle Holocene involved similar but more frequent visits to the area. Given the similarities of the assemblages to other early Holocene sites in the region, a pattern of small scale hunting occupations unfolds in southcentral Alaska. All of these sites, Eroadaway,

Carlo Creek, Panguingue Creek, SRO, and TCO contain small clusters of task specific activities likely undertaken by relatively few individuals engaged in hunting or game processing. Thus, this region was primarily used by small hunting parties undertaking specialized tasks. These groups may have targeted game along the banks of the Susitna

River during winter months at a time before dense spruce forests appeared. Only further investigation and the discovery of more sites can resolve these questions, but until then theoretical models explaining these trends should be developed and tested (Chapter V).

CHAPTER IV

UPLAND LITHIC PROCUREMENTAND HUNTING STRATEGIES

In this chapter, I present a lithic analysis of the assemblages from Bull River II (HEA-

389) and Costello Creek (HEA-393), two important alpine (>914 m asl) tool production

sites in southcentral Alaska. These assemblages are significant for several reasons. First,

because the sites are set just south of Broad Pass, it is reasonable to assume that they

were occupied by people who used the Pass to move between the north and south slopes

of the central Alaska Range. Additionally, because of the upland nature of the area, the

assemblages probably represent prehistoric activities undertaken during the late summer

or early fall and this assumption is used to form the basis of a wider seasonal explanation

for landuse and technological organization in subsequent chapters.

Ethnographic reports suggest high latitude hunters relied on a highly complex

composite technology with many intricate but repairable pieces. These tools required

immense technological knowledge and energy to produce and maintain (Oswalt 1976).

Binford (1980) recognized that arctic hunter-gatherers needed to be highly mobile and

have a complex logistical strategy for hunting because resources are temporally and

spatially clustered on the landscape and separated by great distances. Consequently,

additional planning and reduced opportunity are involved in procurement strategies.

Efforts to survive require a reliable technological system. Abundant archeological assemblages associated with presumably well organized arctic and subarctic hunters frequently consist of thinning and retouch flaking debris, the remnants of maintaining formal tools that were ultimately carried from site to site. Sustained tool maintenance and transport is indicative of a specialized technological organization, and in higher 95

latitudes, it is associated with high mobility and logistical subsistence strategies (Blades

2001:10-11; Kelly 1983; Shott 1986:22; Torrence 1983:13-18). However, retooling sites

represent only a portion of the technological systems employed in prehistoric Alaska.

Other site types have been discovered including lithic quarry, butchering, and ice patch

kill sites. These sites represent lithic reduction facilities and resource procurement sites

essential to the economic system.

Study Area

The Bull River II and Costello Creek sites are located in southeastern Denali National

Park and Preserve, 27.5 km southwest of Cantwell, between the Bull River and Costello

Creek where a complex network of small upland tributaries drain from the mountains

buttressing Easy Pass into the West Fork of the Chulitna River (Fig. 4.1). This region has

both precipitous and gentle peaks that reach in excess of 1370 m (4500 ft) a.s.l. and possesses a seasonal abundance of resources including animals, plants, and toolstone.

Soil formation is generally consistent with approximately 40 to 60 cm of silty-clay soil horizons capping basal layers, but on some hills and ridges deflated blowouts occur.

Vegetation cover is a blend of low shrub alpine tundra consisting of a variety of berries, lichens, and flowers. Dense patches of dwarf birch are found at lower elevations along the major drainages and south facing slopes. Faunal communities include small caribou (Rangifer tarandus) in small herds, grizzly (Ursus arctos) and the occasional black bear (U. americanus), Dall sheep (Ovis dalli), beaver (Castor canadensis), arctic fox (Vulpes fulva), wolf (Canis lupus), hare (Lepus sp.), ground squirrel (Spermophilus parryii), ptarmigan (Lagopus sp.), and in brushy river bottoms moose (Alces alces). No fish species live in the small streams or ponds in this study area. 96

Figure 4.1. Location of the Bull River II (HEA-389) and Costello Creek (HEA-393 sites). Sites mentioned in text are labeled by last three digits of AHRS numbers. Locations of sites are offset by 500 m to preserve integrity. Denali National Park boundary in yellow. Note extensive snow and ice patches in the nearby mountains. Map modified from AlaskaPak for ArcGIS 9.2 from the National Park Service.

Bull River II is situated at the northern end of a 2 km long north-south trending bench at an elevation of 1030.8 m (3382 ft) a.s.l. Paralleling the bench to the west is a 97

narrow beaver pond. The bench also contains at least two additional lithic sites (HEA-

388 and HEA-232), one discovered in 2007 and the other in 1988 (Lynch 1996). Part of

the prehistoric attraction to this bench is the commanding view of the lowlands and easy access to freshwater, toolstone, and upland resources. No ideal occupation point occurs along the entire 2 km long bench that would have attracted prehistoric people repeatedly over time, and hence deposition of palimpsest assemblages has been limited.

Bull River II overlooks a small beaver-dammed tributary of the Camp Creek drainage and is just south of an unnamed western tributary of the Bull River. An old spruce tree about 1.6 m tall is located approximately 110 m south of the site on the western slope of the bench. Costello Creek is located halfway between the headwaters of

Camp Creek, an eastern tributary of Costello Creek, 2.5 km southwest of Bull River II.

The Costello site is positioned in a similar topographic setting at 1038.8 m (3408 ft) a.s.l. but the setting differs in that it is on a small knoll at the base of a larger hill rather than on an elongated bench. Although the age of the Costello Creek site is currently unknown, the attributes and stratigraphic position of its lithic assemblages are comparable to those

at Bull River II.

Research History

In 1988, A. J. Lynch (1996) conducted aerial and limited pedestrian surveys in this area

as a part of a larger park-wide investigation primarily aimed at documenting historic sites near Dunkle Mine. During this survey, three waste flakes, two microblades, and a small

potential microblade core at the unnamed HEA-232 site were identified. The artifacts

were noted on a deflated slope at the southwest end of a bench. Raw material in the

assemblage included black chert, dark gray chert, white chalcedony, and obsidian. A 98

single test pit was excavated but no subsurface artifacts were found and the surface finds were not collected (Griffin 1990:245; Lynch 1996; Saleeby 2000:420).

The HEA-232 site report drew my attention to the area, and in 2007, on behalf of

Denali National Park and Preserve, I conducted a one month intensive subsurface survey of this region with a six person crew comprised of graduate and undergraduate students from the University of Nevada, Reno and the University of Alaska, Anchorage (Wygal

2007c). We focused attention on elevations greater than 914 m (3000 ft) around the

Camp and Costello creeks, beaver ponds, and several unnamed western tributaries of the

Bull River. Ice patch locations were explored in higher elevations >1220 m (>4000 ft) a.s.l. The work yielded 11 newly discovered prehistoric sites, ranging from small lithic surface scatters to dense buried components. No evidence of microblade technology was found; most sites contained biface and bifacial flake fragments made on siltstone.

Obsidian artifacts were recovered at two locations, HEA-232 and the Camp Creek 2 site

(HEA-391). These were sent to J. R. Speakman and N. Slobodina at the Smithsonian

Institution for X-ray fluorescence (XRF) analysis. The results were “a solid match” with

Cook’s (1995:97) group K obsidian from Dry Creek component II (R. J. Speakman 2007 pers. comm.). Most significant of the discoveries were the rich lithic assemblages from

Bull River II and Costello Creek deposited near the base of a buried soil horizon.

The cultural components at both sites are estimated at 375 m2 in size. Three 1x1 m units were excavated at Bull River II and one at Costello Creek. At both sites, artifacts were piece-plotted in three dimensions using metric tape measures and a line level from a central datum. Sediments were initially excavated by 5 cm arbitrary levels with careful 99

documentation of stratigraphic transitions. All sediments were screened through 1/8-in wire mesh.

Stratigraphy and Dating

The stratigraphic assessment from Bull River II and Costello Creek represent the first approximation of the sedimentology of these sites. The sediments were documented during the summer of 2007 during the first phase of our archeological survey of the area and thus may be subject to more detailed field and laboratory analyses in the future. In the following field characterizations, I used the soil guide and texturing flow chart described in the previous chapter (Midwest Geosciences Group n.d.) at the Bull River II site. The same procedures were then independently repeated at the Costello Creek site with similar conclusions. Based on my observations over the course of the month long investigation of this project area, the majority of the region shares strong similarities in sedimentary formations between sites occupying elevated landforms at or around the

1006 m (3300 ft) elevation contour.

Bull River II

Five stratigraphic layers were documented at Bull River II (Figure 4.2), all of which are alluvial deposits, save the basal stratum which is colluvial. The uppermost layer, stratum

5, is a thin vegetation root mat from 0 to 3 cm below the ground surface. The organic horizon (stratum 4) is particularly dark and contains a high concentration of grey silt deposits that grade downward and tend to concentrate toward the bottom of the stratum,

10 to 15 cm below the surface. The grey silts in stratum 4 appear to be a heavily weathered tephra because of a small percentage of angular glass shards observed in the field with a hand lens and the known local occurrence of regional tephra markers, such as 100

the Cantwell ash from the Carlo Creek site (Bowers 1979). These silts marked a distinct

transition to a red/brown silty clay loam deposit (stratum 3b). In some parts of the

project area, particularly at the Camp Creek site (HEA-390), the uppermost levels of

stratum 3b are so heavily oxidized that it forms a crusted barrier which sealed the cultural

component while in other areas, this stratum was more difficult to detect. Below stratum

3b, is the main occupation horizon contained within a distinct paleosol (stratum 3a).

Near the contact of stratum 3a with stratum 2, there is a particularly compact and organic-

rich, black and brown mottled sub-horizon interpreted as the primary occupation surface

most noticeable at a depth of 22 to 27 cm below the surface (ca. 28 to 34 cm below

datum). Stratum 2 is a sandy clay loam horizon and the basal layer (stratum 1) is non-

uniform unsorted colluvium with round

boulders, angular shatter, and fractured

schist mixed in a sandy loam matrix.

Although no faunal elements were

recovered, portions of the original

occupation surface at Bull River II have

remained relatively well preserved for an

alpine site. Other parts of the site have

suffered freeze-thaw disturbances. Artifact intensity increases with depth below surface

while artifact pitch (vertical orientation) decreases with depth. The archeological

material at Bull River II was originally deposited in a single component and vertically oriented artifacts found in the upper levels of stratum 3b were transported upward 101

through cryoturbation. Most artifacts (73% of finds in the excavation block) were

recovered within the primary cultural zone in stratum 3a near its contact with stratum 2.

In three locations within the cultural zone, there were small features of resinous or

greasy black stains associated with in situ artifacts (Fig. 4.3). These features were

collected, wrapped in aluminum foil, and sealed in a ziplock bag. Once out of the field the samples were air dried for 72 hours, then repackaged in fresh foil and bags. In

December of 2008, John McCormack of the Department of Geological Sciences and

Engineering at the University of Nevada, Reno analyzed the sample using a scanning

electron microscope (SEM) equipped with an energy-dispersive detector, and

characterized the elemental composition of the resinous stain. Elemental analysis of

individual particles indicated the sample was comprised of Al, Si, and O in amounts

consistent with standard kaolin-type clay soils but with an abnormal level of carbon.

Figure 4.3. Carbon rich resinous stains at the base of Stratum 3 at Bull River II. Stains are from Unit 2 NW ¼ (right) and SW ¼ (left). Both are from within the cultural horizon.

In a similar study, Kedrowski et al. (2008) isolated the chemical composition

from fatty acids recovered from bone hearth residues at Swan Point, in the Tanana

Valley. Although that level of analysis is beyond the scope of this dissertation, the Swan

Point samples were carbon rich, as are the stains from Bull River II. Because of their 102

association with the cultural zone, these resinous stains are interpreted as human derived, rather than from natural agency. Whether the Bull River II samples resulted from similar

bone burning activities as were undertaken at Swan Point has not been determined;

however, the association of the resinous stains with in situ artifacts strengthens the

argument that portions of an intact occupation surface were left unaffected by

cryoturbation within the lower levels of stratum 3 at Bull River II.

Figure 4.4. Elemental composition of organic stains from the Bull River II occupation horizon.

Charcoal recovered from near the eastern wall of excavation Unit 1 (Fig. 4.5)

occurred in a series of small clusters, some several cm above other charcoal samples but

all associated with undisturbed concentrations of artifacts. Several charcoal fragments

were in excess of 2 cm in diameter, but most fragments measured between 1 and 2 cm.

Four of these individual fragments were dated between 12.5 and 12.2 ka (Table 4.1),

coincident with the Younger Dryas.

The charcoal recovered from Bull River II was the only evidence of charred

material observed in any of the subsurface tests (n = >1000) from this survey region.

Based on my own observations, the internal structure of the charred remains appears 103

consistent with woody plant stalk, not twig or root. Thus, it is not likely the charcoal was

derived from tundra fire or root burn and therefore is most likely associated with the

human occupation of the site.

Figure 4.5. Point provenienced artifacts and charcoal sample from Bull River II.

Table 4.1. Radiocarbon data from the Bull River II site. LAB Number 14C BP δ13C Cal BP 1 Note BETA234748 10,310+50 -26.7 12,180+150 cluster BETA234749 10,350+50 -24.6 12,260+150 cluster BETA234746 10,460+50 -25.4 12,410+150 cluster BETA234747 10,490+50 -26.7 12,460+140 cluster 1 Calibrated using CalPal05 and the Intcal04 curve at one sigma (Weninger and Jöris 2004; Weninger et al. 2005). All dates are AMS on single charcoal fragments.

Costello Creek

The excavation at Costello Creek revealed a similar but deeper stratigraphic profile than at Bull River II (Fig. 4.6). Stratum 5 is relatively thin root mat that terminates near the ground surface. The grey silts (potential tephra) within the organic horizon (stratum 4), which are particularly noticeable at Bull River II, are thin to discontinuous at Costello

Creek. Stratum 3 is comprised of a brown silty clay loam deposit but the oxidation that is so prevalent in the upper layers of this deposit at Bull River II and other sites in the 104

region, was not as noticeable in the Costello Creek

test unit. The majority of artifacts occurred in the

lower levels of stratum 3, a compact sub-horizon positioned just above its contact with the underlying stratum 2. This sub-stratum is interpreted as the occupation surface but no dense concentrations of organic staining or charcoal were recovered at

Costello Creek. Stratum 2 is a sandy clay loam horizon that is thicker than at Bull River II. The basal layer (stratum 1) is sandy loam colluvium with unsorted pebbles and boulders of various sizes. The assemblage remains undated and clearly further investigations are encouraged; however, due to lithic and stratigraphic similarities with

Bull River II it could potentially date to the Younger Dryas period.

As at Bull River II, the Costello Creek assemblage probably represents a single archeological component, although further investigation of the site is encouraged. Other similarities between the sites include the stratigraphic placement of the occupation horizon near the contact of stratum 3 with the underlying stratum 2 as well as cryoturbation processes that resulted in the vertical displacement of artifacts upward in the profile. Further disturbed through bioturbation, Costello Creek also has intact patches of the occupation surface, although these were not as well defined as at Bull River II and no charcoal or features were recovered from Costello Creek. Despite the disturbances, careful documentation of the unit fill distinguished between displaced artifacts and the 105

most probable location of the occupation surface within the profile. Figure 4.7 is a three-

dimensional representation of the 1x1 m excavation unit and point provenienced artifacts

compared to a section of the profile.

Figure 4.7. Stratigraphic profile and block diagram from Costello Creek.

Artifact Assemblages

At both sites, cryoturbation has caused the movement of some artifacts up in the profiles

from the original occupation surface; however, because this surface has been identified in

both sites, the assemblages were analyzed as single archeological components.

Methods

Lithic attribute analyses from the Bull River II and Costello Creek sites followed the same methodology described in Chapter III. Attributes documented for all lithic pieces include raw material type and color, degree of dorsal cortex, platform preparation, and metric data. Tools are distinguished from flaked debris (except for those associated with primary reduction, i.e., cores and core fragments) and are further categorized as formal or informal varieties. Additional traits recorded on tools include edge angle, condition, and the following retouch characteristics: form, degree of invasiveness, location, and number 106

of retouched margins. No blades, microblades, burins, or associated technology were found in the Bull River II and Costello Creek assemblages.

Bull River II

Recovered at Bull River II were 566 pieces of flaked debris and 13 tools (sum = 579 lithic pieces). Raw material at Bull River II consists primarily of blue-grey siltstone

(88.9%). Material types also include sandstone (6.6%), chert (2.8%), basalt (0.7%), claystone (0.7%), and chalcedony (0.3%). All tools were produced from siltstone. Raw material frequencies among flaked debris mirrors the overall assemblage with siltstone as the principal type (88.7%) followed by lesser numbers of sandstone (6.7%), chert (2.8%), claystone (0.7%), basalt 0.7%), and chalcedony (0.4%).

All formal tools at Bull River II are biface fragments (n = 5) consisting of unhafted stage 1, stage 2, and stage 4 biface fragments (Fig. 4.8). Each of these bifaces has scalar retouching except the stage 2 biface, which is stepped. Biface retouch invasiveness ranged from 16.8 to 32.57 cm and edge angles varied between 25° to 38°.

Nearly one in three tools were retouched on a single edge (30.8%), but most (46.2%) have two worked edges, 15.4% have three, and 7.7% have four worked edges.

Informal tools (n = 8) included unretouched utilized flakes (23.1% of entire tool assemblage), a tci-tho (7.7%), and notched flake (7.7%). Retouched flakes tended to be unifacially worked with marginal and scalar retouch forms. The notch measured 11.35 cm on the right lateral margin of a distal flake fragment and bears no other retouch. No hammer or anvil stones were recovered. The tci-tho scraper differs from those recovered from the Susitna lowlands only in that the Bull River specimen has stepped retouch on two lateral margins whereas the Susitna artifacts presented use wear only. 107

Figure 4.8. Tci-tho and biface fragment from Bull River II. (sketch by E. Pellegrini)

Many of the tools at Bull River II are broken (61.5%) and often constructed on

flakes (92.3%). The tci-tho scraper is the only tool with >90% dorsal cortex, an unhafted biface (stage 2) has 51-90% cortex, and the remaining tools (84.6%) have no cortex.

Among flaked debris (n = 566), 75.8% lacked cortex and 17.3% have greater than 90%

cortex. Most of the Bull River II tools were between 5 and 7 cm in size (92.3%) and one

was 7-9 cm. Flaked debris ranged in size from > 1 cm (3%), 1 to 3 cm (57.1%), 3 to 5

cm (28.3%), 5 to 7 cm (9.2%), 7 to 9 cm (1.6%), and 9 to 11 cm (0.9%). 108

Most tools lack an identifiable platform (53.8%). In cases where the platform is present most are simple (38.5%) followed by cortical platforms (7.7%). None are complex or crushed platforms. On flaked debris, platform preparation includes simple

(36.9%), complex (14.5%), cortical (7.1%), and unidentifiable (41.5%) varieties.

Debitage type classifications include complete (29.2% of the debitage assemblage) and fragmented flakes (35.5%) produced primarily from siltstone cobbles. Primary core reduction is represented by angular shatter (4.2%), primary and secondary cortical spalls

(21%), and fragmented multidirectional flake cores (0.9%). When complete flakes are counted (n = 165), flaked debris associated with core reduction equals more than 55% of the debitage assemblage. Biface thinning flakes (7.4%) and retouch chips (1.1%) indicate that biface production was common but more refined retouching activities were less frequent.

Costello Creek

The Costello Creek lithic assemblage contains 423 pieces of flaked debris and 4 tools

(sum = 427 total lithics). Raw material types are primarily siltstone at Costello Creek.

The assemblage is dominated by siltstone (91.8%), primarily a brown to tan variety, with lesser amounts of sandstone (5.2%), basalt (2.3%), and chert (0.7%). Because there are only four tools in the assemblage, raw material frequencies of flaked debris closely mirror the assemblage as a whole: siltstone (92.2%), sandstone (5.2%), basalt (2.1%), and chert (0.5%).

The tool assemblage consists of a side scraper on brown siltstone with stepped retouch on the lateral margins, a retouched flake also on brown siltstone with marginal retouch, a notched flake on tan chert (notch size = 20.38 cm) with an edge angle of 44°, 109

and an unretouched utilized flake on basalt. No hammer or anvil stones were recovered at Costello Creek. All of the tools were manufactured on flakes and only the notched flake was a complete specimen. All of the tools are considered informal because they were expediently manufactured, used, and discarded in a single setting.

The only tool at Costello Creek with dorsal cortex is a retouched flake with 51-

90% surface coverage. Among flaked debris, 84.1% lack cortex and 11.6% have greater than 90% dorsal cortex. Two tools are between 1 and 3 cm and one is 3 to 5 cm and another is 5 to 7 cm. Flaked debris (n = 422) size classes range from <1 cm (1.7%), 1 to

3 cm (64.2%), 3 to 5 cm (25.1%), 5 to 7 cm (6.2%), 7 to 9 cm (2.1%), 9 to 11 cm (0.5%), and 11 to 13 cm (0.2%).

Two of the tools at Costello Creek have simple platforms preserved; platforms are not preserved on the remaining tools. Platform preparation on flaked debris includes simple (37.4%), complex (16.8%), cortical (10.2%), and unidentifiable (35.5%).

Debitage types include complete flakes (33.4%), fragmented flakes (33.2%), biface thinning flakes (13.3%), retouch chips and chip fragments (< 1%), and a blade-like-flake

(0.5%). Detritus from primary reduction activities includes primary (10.9%) and secondary (3.1%) cortical spalls and angular shatter (4.2%). The frequency of debitage types suggests biface production was common at the site and more than half (51.9%) of the flaked debris is consistent with core reduction activities.

Discussion

In both assemblages, raw material types are skewed in favor of siltstone, and tools were recovered in frequencies too low for meaningful statistical analyses. Consequently, 110

descriptive comparisons of the assemblages were deemed most appropriate for intra and inter-site interpretations.

Tools in common between the Bull River II and Costello Creek assemblages include expedient notched flakes and flake tools. While the Bull River II assemblage is dominated by late stage biface fragments (Fig. 4.9), no bifacial fragments were recovered at Costello, although the percentage of biface thinning flakes comprised more than 13% of that assemblage (Fig. 4.10).

Notch Tci-thos Side Scraper Retouched flake Flake tool Biface unhafted

0% 10% 20% 30% 40%

Bull River Costello Creek

Figure 4.9. Frequency of tool types at Bull River and Costello Creek.

Core fragment Worked chert cobble Angular shatter Bifacial thinning flake Retouch chip Retouch chip frag. Sec. Cortical spall Prim. Cortical spall Blade-like-flake Complete flake Flake fragment

0% 10% 20% 30% 40%

Bull River Costello

Figure 4.10. Frequency of flaked debris at Bull River and Costello Creek. 111

Debitage type frequencies are remarkably similar between the sites with relatively

high numbers of complete flakes and cortical spalls, and low numbers of retouch chips

and chip fragments. Multidirectional core fragments from both sites are characterized by

three or more flake scars, with the largest of scars ranging from 23 to 68 cm. Worked

margins on cores varied from a single retouched margin to all margins with flake scars.

These patterns suggest lithic reduction was a primary activity in both sites. High

numbers of bifaces and biface thinning flakes as well as the absence of microblades

indicate these sites were late stage bifacial blank production facilities.

Survey of the Camp Creek tributary directly between Bull River II and Costello

Creek revealed an abundant supply of sedimentary rocks including siltstone, sandstone, and claystone. These materials were exposed in tabular form in cut-banks and as cobbles in Camp Creek. Most common were the blue-grey banded and brown to tan varieties, similar to those recovered in the archeological sites. Other materials were observed in the unnamed tributaries of the Bull River, including quartzite and basalt cobbles; however, these material types were exceedingly hard and difficult to flake, unlike the siltstone which is a more workable material.

Because of the ample supply of siltstone in the region, this material comprises the

bulk of the tools (Fig. 4.11) and flaked debris (Fig. 4.12) from both sites. Smaller

amounts of chert, basalt, and chalcedony also occur among the flaked debris and are

probably non-local in origin as these lithic types did not have primary cortex in the

archeological sites and were not observed naturally in the Bull River-Costello Creek area.

It is important to note that, although XRF analysis has yet to be completed, the two 112

chalcedony flakes recovered from Bull River II are similar in color and texture to those from TCO, and a sample of this lithic type was recovered by Coffman and Goebel

(Coffman 2006 pers. comm.) from natural deposits in Hurricane Gulch, an eastern tributary of the middle Chulitna River ca. 32 km south of Costello Creek and Bull River.

Claystone Siltstone Sandstone Chert Chalcedony Basalt

0% 20% 40% 60% 80% 100%

Bull River Costello Creek

Figure 4.11. Frequency of raw materials among tools at Bull River II and Costello Creek.

Claystone Siltstone Sandstone Chert Chalcedony Basalt

0% 20% 40% 60% 80% 100%

Bull River Costello Creek

Figure 4.12. Frequency of raw materials among flaked debris at Bull River II and Costello Creek.

Flake size and degree of dorsal cortex are indicators of the relative distance (or time) (Chapter III) since lithic procurement (Feder 1980; Newman 1994). Tools at Bull

River II tend between 3 and 7 cm in maximum size and two tools had more than 50% 113

dorsal cortex. There was a similar pattern at Costello Creek with the majority of tools between 3 and 5 cm and a slightly higher percentage of cortex (Fig. 4.13 and Fig. 4.14).

100.0%

50.0%

0.0% 0% 1-10% 10-50% 50-90% >90%

Bull River Costello Creek

Figure 4.13. Degree of dorsal cortex on tools at Bull River II and Costello Creek.

100.0% 80.0% 60.0% 40.0% 20.0% 0.0% 0% 1-10% 10-50% 50-90% >90%

Bull River Costello Creek

Figure 4.14. Degree of dorsal cortex on debitage at Bull River II and Costello Creek.

Degree of dorsal cortex measurements on flaked debris is indicative of core

reduction activities. Relatively high percentages of the debitage from Bull River II and

Costello Creek (15% of the combined total of debitage from both sites) had more than

90% dorsal cortex (Fig. 4.15). Flake size classes (Fig. 4.16) were also consistent between the two sites with most between 1 and 3 cm (BR = 51.7%; CC = 64.2%). Flakes between 114

3 and 5 cm are also prevalent (BR = 28.3%; CC = 25.1%) but debitage pieces less than 1

cm in size were relatively rare (BR = 3%; CC = 1.7%). Consequently, tool production

rather than maintenance activities were undertaken at these sites.

100.0% 80.0% 60.0% 40.0% 20.0% 0.0% < 1 cm 1-3 cm 3-5 cm 5-7 cm 7-9 cm

Bull River Costello Creek

Figure 4.15. Tool size classes (cm) at Bull River II and Costello Creek.

80.0% 60.0% 40.0% 20.0% 0.0% < 1 cm 1 < 1 3 - 5 - 3 cm 3 7 5 cm 5 - 9 7 cm 7 - 9 cm 9 11 - 11 cm 13 - 13 cm - 15 cm

Bull River Costello Creek

Figure 4.16. Debitage size classes (cm) at Bull River II and Costello Creek.

Most tools were made from complete flakes. Platform preparation on tools (Fig.

4.17) is most frequently simple at the sites (BR = 38%; CC = 50%), although many tools

lacked platforms (> 50% at each site). Among flaked debris, platform preparation is uniform between the assemblages, with simple platforms dominating. The relatively 115

common occurrence of cortical platforms in both assemblages (BR = 7.1%; CC = 9.3%)

is an indication of primary core reduction. Also common are complex platforms (BR =

14.5%; CC = 15.4%), which typically occur on bifacial thinning flakes (Fig. 4.18).

60.0% 40.0% 20.0% 0.0% Cortical Simple Complex Unidentifiable

Bull River Costello Creek

Figure 4.17. Tool platform preparation at Bull River II and Costello Creek.

50.0%

0.0% Cortical Simple Complex Unidentifiable

Bull River Costello Creek

Figure 4.18. Platform preparation on flaked debris at Bull River and Costello Creek.

The density of the deposits and attribute frequencies described for both sites

support the hypothesis that lithic procurement and initial reduction to biface production

of locally available siltstone were the primary activities undertaken at the Bull River II and Costello Creek sites. Moreover, raw material was amply available and thus conservative practices like tool recycling and microblade production were unnecessary. I 116

conclude that, given the nature of the assemblages, the production and curation of bifacial blanks and projectile points was a significant activity in these alpine settings.

Upland Kill Sites

Most early prehistoric sites in central and southcentral Alaska are known from relatively large sites in the Nenana and Tanana valleys. The Bull River II site is the earliest dated occupation in Alaska above 3000 ft and therefore it represents a significant deviation from this pattern. In Beringia, the Pleistocene-Holocene boundary is arguably the best documented archeological period; yet, four recent syntheses (Bonnichsen and Turnmire

1999; West 1996d; Wygal 2003; Yesner 1996) of sites dated between 13 and 8 ka provide little detail on kill sites. These reports document an abundance of overlook and retooling sites in eastern Beringia and there are many examples of projectile point bases, burins, cores, and other implements suggesting the discard and repair of broken hunting implements in many eastern Beringian sites. For example, in northwest Alaska, several early Holocene sites related to caribou hunting are positioned directly adjacent to portages or migration pathways but contain assemblages more characteristic of retooling stations (Anderson 1968, 1988) or quarry and biface production activities (Rasic 2003).

Although rare, cave sites have yielded organic artifacts still hafted with stone tools in

Alaska (Ackerman 1996d:470; Larsen 1968; Schaaf 1988), but the attribution of these sites as human occupations has been debated (Hamilton and Goebel 1999; Satler 1997;

Vinson 1988, 1993). None of these are primary kill sites but they emphasize the necessity of tool production and repair in prehistoric hunting strategies.

Within the past 10 years, a number of researchers have investigated ice patches in

Alaska and the Yukon Territory (Dixon et al. 2005; Hare et al. 2004; Helwig et al. 2008; 117

VanderHoek et al. 2007a, VanderHoek et al. 2007b; Wygal 2008b). These site types primarily contain killing implements and faunal elements without significant amounts of marrow or meat (primarily crania, mandibles, and lower limb bones). They also have not produced proximal point fragments or other evidence of tool repair (i.e., burins, hammerstones, or cores), suggesting upland ice patches were a principal location for direct game procurement and not related to tool production or retooling endeavors.

However, before assessing ice patch sites in eastern Beringia, it is useful to briefly review prehistoric kill sites from other areas of North America because these may provide some context for potential prehistoric hunting strategies in southcentral Alaska.

Analogous Prehistoric Hunting Sites and Strategies

Analyses of animal butchering from kill sites in the contiguous United States suggest a broad repertoire of prehistoric hunting behaviors. Some archeologists claim that butchering patterns and faunal elements in archeological contexts represent preferential meat cuts (Frison et al. 1976:48, Kehoe and Kehoe 1960:423), but these may also be an indicator of transport distance from the kill to consumption sites (White 1954:256). In many kill sites there appears to be a common pattern of leaving cranial, maxillary, mandibular, and vertebral elements at the kill location (Kehoe and Kehoe 1960; White

1954). For example, at the Hawken site in Wyoming, Frison et al. (1976) reported this trend but added that most marrow producing bones had been removed with the exception of some broken in a fashion suggesting marrow extraction had occurred at that location.

Non-meat bearing elements such as phalanges and metatarsals display no evidence of use and their presence suggests the hooves were separated from limb elements prior to transport. Additionally, mandibles frequently bore evidence of cut-marks consistent with 118

tongue removal (Frison et al. 1976:50-51). These finds are interesting because, aside

from caribou antlers, mandibles and cranial elements are the most frequent faunal

element observed in Alaskan ice patch settings, although it should be noted that these elements are also frequently left behind in animal predation sites (Haynes 1993).

Alpine hunting sites are probably the best analogs for Beringian kill sites. Several reports of high altitude (>2000 m) hunting sites from the Shoshone Mountains, near

Yellowstone National Park, have provided organic artifacts related to sheep hunting. The

Yellowstone sites contained cord twisted nets and a wooden club. Hunting facilities such as drive lines, hunting blinds, and cliff jumps have been found at several sites in this region and range in age from late Paleoindian to historic times (Benedict 1996; Frison et al. 1986, 2004:165-167; Hogan 1974; Keyser 1974). In the central Alaska Range, the ice patch sites discovered in the Tangle Lakes area (VanderHoek et al. 2007a; VanderHoek et al. 2007b) suggest similar tactics were involved in upland hunting, and meat caches have been found in upland settings of Denali National Park (Eldridge 2008); however, support facilities such as hunting blinds or tool production sites have not yet been associated with ice patch hunting in Alaska. To fully understand alpine hunting behavior in southcentral Alaska, archeologists need to investigate the full range of site types in higher elevations. These include tool production sites like Bull River II and Costello

Creek where technological preparations for upland hunting were undertaken.

Ice Patch Kill Sites in Alaska

The Alaska State Office of History and Archeology (OHA) has been monitoring melting ice patches in the Amphitheater Mountains, Tangle Lakes Archeological District north of the Denali Highway and east of Cantwell. The region has provided a glimpse of the tools 119

and strategy involved in upland resource procurement (VanderHoek et al. 2007a). In

2003, a barbed antler point, twisted sinew on an arrow shaft, three refitting sections of an

arrow shaft (52 cm in length), and associated feather fletching, were recovered by OHA

archeologists. Projectile points and a wooden foreshaft (Fig. 4.19) were recovered from

the Basalt Lake ice patches by VanderHoek, Holmes, and Wygal in 2004. In that same

year, a 75 cm long wooden stave, notched on its distal end and interpreted as a ground

squirrel snare, was recovered by VanderHoek, Wygal, and Tedor in the Delta River Ice

Patches (VanderHoek et al. 2005; VanderHoek et al. 2007a).

Figure 4.19. Projectile points and wooden foreshaft from Basalt Lake Ice Patch 4 (VanderHoek et al. 2007a:191) (photo by B.T. Wygal).

None of OHA ice patch finds described above predate AD 900 (VanderHoek et al.

2007a) leading to the interpretation that the oldest cultural layers investigated thus far in 120

the Tangle Lakes ice patches are only 1.1 ka. Older deposits occur in the southern Yukon where an antler projectile point with an incised slot and a hafted microblade fragment were recovered and dated to 7310+40 BP / 8110±50 cal BP (Helwig et al. 2008).

The nature of artifact finds from high elevations in the central Alaska Range suggests logistical mobility, curated technology, and targeted hunting strategies. Hunters likely operated in high altitudes in small specialized groups tasked with particular purposes like the planned killing of caribou at predetermined locations. A preponderance of young adult or near-adult caribou crania with intact antlers have been observed in the

Tangle Lakes and Denali ice patches. The presence of these elements and the lack of long bones in ice patches could be an indication of hunting and primary processing rather than natural animal deaths, although bear and wolves are also known to hunt these areas.

However, the significance of crania with intact antlers supports the hypothesis that ice patch ambushes were best orchestrated in late summer and early fall when snow melt was at its greatest extent and animals congregated in fewer locales. Caribou are a predictable species (Kenyon 1997:3), making ice patch localities ideal for targeted hunting strategies.

In terms of hunting tactics, the location of ice patches that have yielded artifacts tend to be linear or horizontal snow fields near the very top of east west trending ridges, where hunters could sit unseen and out of the wind, and where they could ambush caribou herds arriving at the snow fields from below.

In the 2004, I participated in an additional OHA survey administered by R.

VanderHoek in the alpine setting south of Landmark Gap, Tangle Lakes Archeological

District (VanderHoek 2005). This area is 975 to 1158 m (3200 to 3800 ft) asl and has one of the highest concentrations of early Holocene sites in Alaska (Mobley and Morris 121

1981; VanderHoek 2005; West 1967, 1972, 1973, 1974, 1975, 1981, 1984; West et al.

1996a; West et al. 1996b; West et al. 1996c; Zink and Zink 1976). The Landmark Gap

area was heavily utilized by prehistoric hunters, in part because there is a wide variety of

quality chert and argillite outcrops in the vicinity but also because many ideal ice patch

locations are easily accessible in the mountains above this area. During the 2004 survey,

VanderHoek and I revisited many of the previously documented sites (n=23), recorded

new lithic scatters (n=10), and conducted test excavations at the Landmark Gap Trail site

(VanderHoek 2005:17). Our work at the Landmark Gap Trail site followed previous excavations by Mobley (1982) and Gillispie (1992). None of these sites contained

microblades or associated byproducts. Instead, the surface sites in this area contained

relatively high frequencies of bifacial thinning flakes, and the Landmark Gap Trail

excavations yielded a variety of late stage bifacial blanks and related detritus

(VanderHoek et al. 2005).

The potential for these types of finds to occur in the Bull River-Costello project

area is high because of its mountainous environment and prehistoric availability of

caribou, bison, and sheep during summer and fall months. The 2007 survey resulted in

the documentation of high potential ice patch localities in the peaks immediately above

Bull River and Costello Creek (Fig. 4.20); however, due to high amounts of annual snow

fall and low annual temperatures in this area, these ice patches have not yet begun to

erode (Wygal 2008b). Nevertheless, caribou bone elements of the types frequently

discarded by prehistoric hunters at known kill sites have been found in these places (e.g.,

crania with articulated antlers and mandibles) and the sites are suspected to have been a

main attraction for prehistoric hunters. Consequently, the purpose of intensive biface 122

production facilities at Bull River II and Costello Creek, within sight of high probability ice patches, was likely to produce an abundant supply of projectile points for the targeted hunting of caribou and other species in these predictable locations. The site type, location, and emphasis on biface production are consistent with assemblages necessary for prehistoric hunting in uplands.

Figure 4.20. Ice patch above Bull River II and Costello Creek in Denali National Park and Preserve. To date, no archeological sites have been located in ice patch settings in Denali.

Conclusions

Survey and test excavations in the mountains south of Broad Pass, Denali National Park and Preserve, revealed new perspectives on the prehistoric occupation of the upland areas of southcentral Alaska. Significant among these are the Bull River II and Costello Creek buried lithic production sites. Although limited testing has been undertaken thus far, the relatively dense deposits have yielded sizable lithic assemblages. Qualitative 123

comparisons of the sites indicate similarities between raw material selection and flaked

debris suggesting lithic reduction and biface production were the primary activities at

both locations. An intact cultural horizon at Bull River II yielded highly organic resinous

stains associated with artifacts and several small clusters of charcoal. AMS radiocarbon

dates on four individual fragments of charred plant remains yielded an age between 12.5

and 12.3 ka.

The Bull River II and Costello Creek sites are situated in the alpine zone just beneath a series of peaks. The site’s setting is ideal for preparing fall hunts for upland sheep and caribou. Ice patch kill sites from the Tangle Lakes and Yukon suggest alpine hunting was commonly practiced by prehistoric people who occupied the interior mountains of eastern Beringia. Hunting strategies included the targeted hunting of specific species in places where game could be found predictably. Ice patches were ideally suited for this type of strategy because in late fall these locations attract a variety of cold adapted animal species seeking relief from heat and bugs. This type of subsistence strategy requires a level of toolkit preparedness allowing hunters to produce enough armatures to maximize productivity during the times when animal resources are concentrated in these predictable locations. The Bull River II and Costello Creek sites represent tool production facilities related to these upland hunting activities. 124

CHAPTER V

THE MICROBLADE/NON-MICROBLADE DICHOTOMY:

THE EFFECT OF CLIMATE AND SEASON ON TOOLKIT VARIABILITY

The earliest lithic technology in eastern Beringia involved the systematic production of microblades, a trait shared with Siberian progenitors. Since that time, microblade industries have persisted throughout the Holocene and across many technological traditions, but they do not occur in every eastern Beringian site. Understanding why microblade production appears in some sites but not in others is integral to Alaskan prehistory. Climate, raw material availability, culture, and site specific activities have all been offered as mitigating variables, making it difficult to determine why these tools are present or absent in any given site. This chapter presents two models that examine the role of microblade technology in eastern Beringia. First is an analysis of the abundant radiocarbon record, coupled with the paleoclimatic data presented in Chapter II, making it possible to examine the effect of climate change on the frequencies of microblade bearing components through time. Second is a seasonal approach to technological organization developed from known landuse strategies and the beneficial attributes of microblade composite technology that make it superior to stone bifaces as a specialized cold weather implement.

In eastern Beringia, microblades have been found alongside nearly all late

Pleistocene diagnostic artifacts including fluted points, Mesa (Bever 2001a, Bever 2008),

Chindadn (Cook 1969), and notched Northern Archaic points (Esdale 2009; Wygal

2003:65), as well as a variety of other point types (Hoffecker in press). Consequently, the greatest problem in explaining this dichotomy is assemblage variability. Although some 125

researchers have argued sample size could account for many interpretive problems for archeology in general (Rhode 1988), others (Mason et al. 2001:531) have argued that substantially increasing excavation areas would not necessarily resolve variability issues simply because the number of artifacts in any given site varies dramatically irrespective of excavation size.

Although the presence or absence of microblades in sites is the result of a variety of factors, these factors are not infinite and thus can be analyzed systematically. To accomplish this, proposed hypotheses must be testable, which in this case may be summarized in either of two ways: microblade composite technology was a reliable and flexible industry during climatically unstable times and/or it was a seasonally employed subset of a wider repertoire of tools.

As a first approximation for explaining the microblade/non-microblade dichotomy in eastern Beringia, this analysis focuses on assessing two variables related to microblade composite weapons as a cold weather specialty. First is a long-term assessment of the radiocarbon record, with respect to the presence or absence of microblades in well-dated assemblages juxtaposed with worldwide climatic oscillations documented in Alaska and the 18O proxy data from the Greenland ice core. This approach assumes a direct relationship between occupation frequency and population levels (Buck and Bard 2007).

Interesting trends exist with distinct population nadirs coincident with the onset of major ecological or climatic shifts and the primary use of microblade technology at the onset of those crises. Most significant was the total absence of archeological components at the beginning of the Younger Dryas and the dominance of microblade assemblages that followed. The results suggest that although microblade production was persistent at 126

various degrees over the past 14,300 years in eastern Beringia, only those groups who employed it may have survived some of the most extreme changes in climate and ecology.

In a finer grained approach, seasonality is explored as a model for differential tool use on the landscape. Both models link cold oscillations, on both an annual and long term scale, to a greater reliance on microblade technology.

Microblade Technology

By definition microblades are standardized elongate blades with parallel lateral margins, typically less than 20 mm in length and 5 mm in width. They have dorsal ridges or facet scars, remnants of blades previously removed (Flenniken 1987; Goebel et al. 2000;

Kobayashi 1970; Morlan 1976). A wide variety of blade and microblade technologies are the result of entirely different production systems. Consequently, criteria for defining microblades have not been standardized, with some researchers accepting dimensions as large as 30 mm long and 10 mm wide (Collins 1999:10; Owen 1988:2). Microblades, as opposed to blades or macroblades, were deliberately constructed for the purpose of hafting into composite tools. This aspect is reflected by the end product’s small size and standard form (Kuhn and Elston 2002:2), and it is by these measures that archeologists ultimately define the technology.

Microblade technologies "never comprise an entire lithic industry, but are used in conjunction" (Kuhn and Elston 2002:5) with other tool kits and systems that require dependable and stable implements where redundant tasks are frequent (Bleed 1986). For example, it is essential to maintain a sharp cutting edge on a reusable tool while hunting herd animals such as bison and caribou so that multiple animals can be taken without weapon malfunctions or the need for reloading (Kuhn and Elston 2002:5). In cold 127

environments, osseous projectiles inset with microblades are better suited for these tasks than spears tipped with lanceolate projectile points (Elston and Brantingham 2002).

Microblades appear in southcentral Siberia a few thousand years prior to arriving in eastern Beringia and may have developed during the harsh conditions of the last glacial maximum sometime between ca. 22 and 20 ka (Goebel 1999; Goebel et al. 2000:567-568;

Graf 2008). Perhaps the invention of microblade composite tools was a prerequisite for the successful colonization of Beringia and, therefore, the New World (Graf 2008).

Core and Blade Technologies in Eastern Beringia

How we interpret the expansion of late Pleistocene technologies into eastern Beringia is dependent on an accurate assessment of technological organization, and these events have ramifications for subsequent migrations into all other areas of eastern Beringia, a process that may not have been complete until the middle Holocene in some cases. The extent to which any of the earliest complexes now defined in eastern Beringia can be clearly categorized into distinct technological complexes remains unclear, primarily because the presence or absence of microblade technology has yet to be adequately explained.

Although disagreements can be found over nearly all of the early lithic industries in eastern Beringia, much of the debate centers on interior Alaska where the Nenana and

Denali complexes are regarded by some as aspects of the same prehistoric tool kits representing a greater Beringian technological tradition (Holmes 2001; Holmes and Crass

2003; West 1967, 1975, 1981, 1996a). Others maintain the Nenana complex should be regarded as a different industry representing an initial peopling of Alaska by foragers who lacked a microblade industry (Goebel et al. 1991; Powers and Hoffecker 1989).

Reconciling these differences and reaching consensus on a comprehensive understanding 128

of the early peopling of eastern Beringia has yet to be achieved (Bever 2001b; Hamilton and Goebel 1999:184; Hoffecker 2001:149).

Some points of agreement include the diagnostic attributes of the Nenana complex as originally defined in the Nenana valley (Powers and Hoffecker 1989). The Nenana complex is an assemblage based primarily on a core and blade (not microblade) reduction strategy and characteristic tear-drop to triangular-shaped bifacial projectile points or knives, large retouched blades, and flake tools such as end scrapers, side scrapers, gravers, and cobble tools including planes and chopping implements (Goebel et al. 1991). For some archeologists (Carlson 1991; Goebel et al. 1991; Goebel 2004; Haynes 2002; Yesner

1996), the Nenana complex was considered a likely ancestor for Clovis, the earliest well- dated Paleoindian complex in the Americas.

West (1967, 1975, 1981) originally defined the Denali complex as those assemblages that contained microblades from multifaceted wedge-shaped cores, specially prepared cores for large blade production, bifacial biconvex knives, flat-topped end scrapers constructed from thick flakes, burins, and burin spalls. Also associated with the

Denali complex as a byproduct of blade production were core tablets created when fresh platform surfaces were generated for blade removal. These distinct byproducts were produced when the "spall… hinged up about one-fourth to one-third of the distance back from the face. A stop notch on the top at that point ensured the termination of the core tablet spall there and prevented the entire core top from being carried away" (West

1996e:303).

Inconsistencies persisted between the separation of the Nenana and Denali complexes because microblades are absent in the Nenana complex but also because the 129

diagnostic Nenana valley points were reminiscent of Chindadn points from Healy Lake in the Tanana valley. At Healy Lake, Cook (1969) argued for the association of the triangular shaped Chindadn points and microblade technology; however, the stratigraphic association was unclear and the site remained an anomaly. The discovery of a “Dyuktai-like” microblade industry at Swan Point CZ 4, dated considerably older than the earliest Denali complex site, at 14.3 ka (12,300 BP), re-opened the debate regarding the role of microblades in the colonization of eastern Beringia (Holmes and Crass 2003; Holmes in press). The debate was further complicated by finds in CZ 3 at the Broken Mammoth site in the Tanana valley, where a small number of microblades (n = 44), but apparently no microblade cores, were found along with small Chindadn-like points. The finds were made within a series of “paleosol stringers” dated by hearth smears between ca. 12.7 and

12.2 ka (Krasinski 2005; Krasinski and Yesner 2008). Discoveries at Swan Point CZ 3 are similar in age (ca. 12.6 to 11.5 ka) and also contain a small number of microblades along with Chindadn-like triangular points, but no microblade cores (Holmes in press). Thus, because the early cultural history in central Alaska has seemingly grown more complex as researchers add to the archeological record, a variety of new perspectives have developed to explain technological variability in Beringia (Goebel and Buvit in press).

In a functional approach, Ackerman (in press) presented intriguing evidence from southwestern Alaska indicating Denali complex microblades were inset in arrow points in addition to spearheads. If correct, that would push the known use of the bow and arrow in

North America back by nearly 10 ka (Dixon in press). Other researchers have inquired rather intensively on the distribution of microblade technology over time (Bever 2006;

Mason et al. 2001; Potter 2008a, 2008b, 2008c; Wygal 2007a, 2007b, 2008a, in press). 130

These analyses attempted to unravel the question of when and where microblades were used on a regional scale in central Alaska, and to some extent southcentral Alaska, but the results are far from conclusive. Most review or evaluate the radiocarbon record and extrapolate a relative measure of population size based on the number of dated archeological components on the landscape. Among the significant findings in the earliest of these studies (Mason et al. 2001) was the apparent overall success of the Denali complex during the relatively cold Mesoglacial period. However, during the more severe

Younger Dryas cold period, the reverse appears to be true with some consensus for a significant reduction and subsequent shift in technological organization coincident with the onset of the Younger Dryas (Bever 2006; Potter 2008a, 2008b, 2008c; Wygal 2007a,

2007b, 2008a, in press).

Potter (2005:571, Table 8.12) compiled data from archeological components within the Tanana River basin dated between 13.9 and 7.8 ka (12,000 and 7000 14C BP). After averaging the radiocarbon data, he compared the presence and absence of microblades and applied a series of Pearson χ2 statistics to detect areas of significance between time intervals and the presence of microblades. Each test indicated no statistical differences.

Potter (2005:572-573) concluded “[t]he apparent pattern of microblade and non- microblade component time differences is not substantiated” possibly due to sampling error. Mason et al. (2001:542) mentioned sampling error as a possibility for the microblade versus non-microblade dichotomy, but favored a climatic explanation instead.

Although Potter (2005:572-603) made many other insightful observations about the presence and absence of microblades in interior Alaska, including proximity to known toolstone sources, mobility, and technological organization, by averaging the radiocarbon 131

data, he blurred what are statistical probabilities. In later works, Potter (2008a, 2008b,

2008c) continued to seek correlations between site elevations and tool types but with a refined methodology and the effect of climate change as a significant factor.

In the following section of this chapter, I present a similar attempt at correlating dated human occupations in eastern Beringia with known paleoenvironmental reconstructions described in Chapter II.

The Radiocarbon Record and Method of Analysis

While the radiocarbon method is often indiscriminately utilized by archeologists, not all dates are equally reliable. Accurate chronologies are essential to interpreting when microblades were used, and to correlate this to paleoclimate. However, archeologists seldom detail why some radiocarbon dates are favored over others. A number of factors influence the reliability of radiocarbon dates including laboratory procedures, association of samples with archeological material, and the type of material being dated (Hedges and

Van Klinken 1992; Pettitt et al. 2003; Stafford et al. 1991). Radiocarbon signatures are often subject to contamination (Haynes 2002:13), especially when recovered from depths of less than one meter (Potter 2005). The impact of nitrogen on the rapid decay of 14C is unclear (Neff et al. 2002), especially as nitrogen fixation in the northern forests is poorly understood (Deluca et al. 2002). Since 1950, greenhouse pollution and atomic detonations have further contaminated the radiocarbon record through ground water seepage (Potter

2005). Thus, the radiocarbon record requires high interpretational maintenance before age estimates are proposed for a given occupation (Graf 2005). Calibration into calendar years is required to allow temporal comparisons of the local manifestations of global events such as the Younger Dryas cold interval (Fiedel 1999; Stuiver and Reimer 1993). Although 132

similar evaluative methods exist (Graf 2005, 2008; MacPhee et al. 2002; Mead and

Meltzer 1984; Pettitt et al. 2003; Rick 1987; Waterbolk 1971), one was developed specifically for this analysis.

Methods of Evaluation

Recent evaluation methods (Pettitt et al. 2003; Graf 2005, 2008) developed a point-based system to assess dates. For instance, given the problems associated with radiocarbon dating, only artifacts directly associated with permanently lined fire hearths or organic artifacts are considered ideal for establishing completely reliable ages (Pettitt et al. 2003); however, relying only on such finds provides a limited sample size in eastern Beringia and it is not unreasonable to assume that where there were arctic foragers, there often were small fires. In cases where dispersed charcoal is mixed among cultural debris and evidence of localized forest fire horizons is lacking, the charcoal likely arrived there through human action. Therefore, well stratified sites with distinct stratigraphic layers and dispersed charcoal that date sequentially in a logical order have been included in the array of accepted dates.

While I utilized similar criteria to critically assess dates from eastern Beringia, many types of information such as methods of analysis were frequently unavailable.

Archeological context, material dated, taphonomy, and date precision were considered in this analysis; however, instead of a point based system, this analysis rejected dates if any one of the rejection criteria were met. The aim was to assign only the most appropriate single radiocarbon age to components within a 100 year interval. Components that contained multiple dates spanning more than 100 years were assumed to have been repeatedly occupied. Two methodological criteria were applied to assess uncalibrated 133

dates, and the results were designated Method I and II datasets. Method I rejected dates if

1) they were not obtained from cultural horizons; 2) if the date was run on soil organics, humates, or combined samples; 3) the dates had standard deviations in excess of 250 years at one sigma; or 4) if the date was from a single cultural feature and did not overlap other dates from that same feature at one sigma. Dates rejected by the Method I criteria are noted in Table 7.9 in the Appendix by the superscript a.

Method II was more conservative and rejected the same dates as Method I but additionally refined the dataset by eliminating dates with a standard deviation of more than

150 years. Dates rejected by Method II are noted by the superscript b. Additional criteria were applied during the evaluation process. Dates considered aberrant by the original investigators (indicated by an asterisk) were rejected based on the site summaries by

Hamilton and Goebel (1999). Finally, in cases where components were represented by multiple radiocarbon samples within a 100 year interval (e.g. 9800-9900), the date with the smallest standard deviation was selected and remaining dates were rejected. Rejection based on this criterion is noted in Table 7.9 by the superscript c. Choosing a one-sigma precision of less than 100 years was not feasible because until recently, radiocarbon estimates rarely had standard deviations that small. Criterion c was applied to both

Method I and II approaches.

In an example of criterion c, the Mesa site contained many hearth features as evidence of multiple occupation events and more than 50 radiocarbon dates have been run from these features (Kunz et al. 2003). The majority of these span an approximately 400- year period. To ensure each potential occupation was represented equally, four dates with the lowest available standard deviations were accepted to represent each of the four 134

consecutive 100-year occupation intervals from the different loci at a site. In no case were weighted means employed to avoid further blurring data that are essentially statistical probabilities. Those dates removed by this criterion were not necessarily anomalous, but were necessary to cull from the data to establish the single most likely age for a given occupation.

Results

The radiocarbon record for Alaska consists of relatively few dates from a limited number of cultural components (Mason et al. 2001). Table 7.9 lists radiocarbon dates (n = 274) from archeological sites in the heart of eastern Beringia, as well as the material dated and age calibrations among other information. Specific attention was given to sites in the

Brooks Range, interior and southcentral regions of Alaska, in addition to sites in the western Yukon Territory. The Aleutians and southeast Alaska were omitted from the analysis because of radically different ecologies, marine carbon reservoir, and potential driftwood effect. Although not every site in eastern Beringia was included in this analysis, those that were form a representative sample for the area. Included for each date is a notation on the material dated and other significant associations. The results produced two datasets of systematically evaluated dates; each is assumed to reflect a single specific occupation event within a one hundred year period. The dual approach allows for a comparison of evaluation methods. All dates were calibrated using the CalPal software

(Weninger 1984; Weninger et al. 2005) at one standard deviation (σ) with the Intcal04 curve (Reimer et al. 2004). These were combined with the presence-absence of microblades to form the primary unit of analysis. 135

Results appear as stacked bar charts in Figure 5.1. Only dates lacking a superscript notation and those labeled b in Table 7.9 were used in the less conservative Method I dataset. Method II differed only in that it had a more restricted σ tolerance by excluding dates labeled b. Each date was assigned to a 500-year interval based on its medium point because it was assumed that errors made by cases where a date’s margin of error overlapped a 500-year interval would become averaged without seriously biasing the overall distribution of dates.

Percentages of accepted dates representing each occupation event are separated in stacked bar format based on the presence or absence of microblades for each 500-year interval. The stacked sections that appear as blue segments represent occupations where microblades have been found and these are juxtaposed with occupations that lacked the technology depicted in green. By combining these datasets with worldwide climatic oscillations (Chapter II) and the GISP2 18O temperature record (Weninger et al. 2005), a hypothetical depiction of technology and component frequency over time is achieved.

Whereas each accepted date approximates the age of a single occupation event, the data compiled in graph format are representative of the frequency of occupation events through time.

Some archeologists interrupt trends from similar types of data as representative of population levels, assuming that more people leave behind more sites (Graf 2008; Mason et al. 2001; Rick 1987; Surovell et al. 2005). Other researchers have applied similar concepts to paleontological studies (Buck and Bard 2007; Kuzmin 2008; MacPhee et al.

2002; Ugan and Byers 2007) but recently such approaches have been questioned. Surovell and Brantingham (2007) found that taphonomic factors skew the distribution of dated 136

components on the landscape towards younger sites. More specifically, issues may result in preservation biases associated with differential geomorphological processes, rate or probability of burial overtime, as well as site preservation and discovery biases (Mason et al. 2001; Surovell and Brantingham 2007). Another issue is the overwhelming researcher bias toward older sites in Alaska. Regardless of the difficulties in reconstructing past populations dynamics, the number and effectiveness of these studies remains on the rise in archeology. I argue that the frequency of human occupation events and the frequency of microblade industries through time are considered a proxy that can be used to recognize prehistoric population and technological responses to climate change.

Distributions in Figure 5.1 are interesting for several reasons. Initial observation suggests Method II yielded a distribution similar to Method I and the skewed effect predicted by Surovell and Brantingham (2007) did not materialize for this dataset. In terms of technology change over time, microblades persisted in various degrees throughout prehistory and increased in significance around 11 ka, coincident with the expansion of the northern birch forests (Bigelow and Edwards 2001:208). Low component frequencies appear to correlate with significant cold and drastic paleoecological events (i.e. a disappearing mammoth steppe, Younger Dryas).

Surovell and Brantingham (2007) suggested focusing on narrower time intervals to minimize distribution biases. In response, I removed all occupations dated after 7.7 ka from the data and narrowed the time intervals to 250 year segments revealing a slightly more complex pattern in Figure 5.2. The finer grained approach more clearly demonstrates the human response to specific climatic oscillations including the particularly severe human response to the onset of the Younger Dryas. 137

Figure 5.1. Microblade/non-microblade dichotomy and component frequency in eastern Beringia (1000-14,500 cal BP). Results of the two-method approach for evaluating radiocarbon data are compared with trends in the presence or absence of microblades and human occupation events arranged in 500-year increments. Climatic episodes are based on the central Alaskan record cited in Chapter II and abbreviated as follows: Medieval Cooling (MC), Neoglacial (NG), younger Younger Dryas (yYD), Milankovitch Thermal Maximum (MTM), Younger Dryas (YD), Allerød (A) and the Older Dryas (OD). GISP2 reflects the Greenland Ice Sheet Delta 18O percent, and was produced using CalPal Climate Proxy Timescale Composer software (Weninger et al. 2005).

138

Figure 5.2. Microblade/non-microblade dichotomy and component frequency in eastern Beringia (7700-14,500 cal BP). Results of the two-method criteria for evaluating radiocarbon data are compared with occupation events dated earlier than 7.7 ka and arranged in 250-year increments. Climatic episodes are based on the central Alaskan record cited in Chapter II. References and abbreviations appear in Fig. 5.1.

Prehistoric Populations and the Presence/Absence of Microblades

The initial colonization of eastern Beringia around 14.3 ka is apparent in Figures 5.1 and

5.2 with microblades in use at a single site, Swan Point CZ 4 (Holmes in press; Holmes 139

and Crass 2003), which may have been occupied during the brief Older Dryas cold episode. During the middle to late Allerød warm period, the frequency of occupations is relatively high and microblade industries are less common at sites in the Tanana valley

(CZ 4 at Broken Mammoth and Mead), and in the Nenana complex occupations (Dry

Creek, Moose Creek, Owl Ridge, and Walker Road). One possible explanation for this trend is that a second and much larger wave of colonizers using a fundamentally different technology arrived after 13.5 ka. Alternatively, postglacial warming may have fostered a substantial population explosion following initial colonization. This was probably the case as people living in the now flooded areas of the Bering Strait were pushed east (Yesner

1996, 2001). Population increases in the interior may have initially forced human groups to expand into peripheral zones where large animal densities are thought to have been higher. Similar events occurred in Upper Paleolithic Europe at that time (Pettitt 2005:163) and in southern Siberia following the LGM (Goebel 1999; Graf 2008).

Did the onset of cold climatic episodes during the late glacial effect the distribution of microblade industries through time? Mason et al. (2001) suggested that during colder climates, hunting and mobility were easier in the arctic and thus fostered an increased population among users of the Denali industry. Their study also revealed a decrease in

Denali occupations during the Milankovitch warm period suggesting increased aridity was detrimental to microblade-using foragers (Bigelow and Powers 2001; Mason et al 2001).

My analysis supports these findings when comparing the number of microblade components dated to the MTM to the preceding 1000 years. The overall trend suggests significant and rapid reductions in occupations were primarily coincident with the onset of worldwide cold periods, including the Older Dryas, Younger Dryas, Mesoglacial, and 140

several Neoglacial periods. Diminishing occupation events during the Milankovitch warm and dry period appear to have been more gradual, with the lowest numbers of occupations occurring toward the end of that period rather than with its onset; however, this trend could also be attributed to the continued spread of the boreal forest.

How did the developing Preboreal and Boreal vegetation stages relate to the use of microblade technology? One clear trend highlighted in this analysis was the dominance of microblade industries around 11 to 10 ka, coincident with the spread of birch (Mason et al.

2001) during the Preboreal stage. Microblade-bearing components remain frequent throughout the Boreal era until approximately 800 years ago when copper trade spread from the Wrangell Mountains.

In most instances, where occupation events are at their lowest, only microblade industries persist and these events correspond with the onset of significant cold episodes and ecological changes such as the Preboreal and Boreal stages. Although abrupt climatic and ecologic changes may have negatively impacted the number of human occupations on the landscape, groups adjusted rapidly, generally rebounding within a period of 500 years.

The most significant nadir occurred with the sudden onset of the Younger Dryas cold event

(Wygal 2007a, 2007b, 2008a, Wygal and Krasinski 2006). Figure 5.2 suggests the crisis was focused between 12,950 and 12,700 cal BP when human occupations are undetectable archeologically. Following the sharp decline, there is a brief resurgence in microblade industries; however, as the Younger Dryas proceeded, the frequency of dated components lacking microblades rises sharply, reaching its highest point immediately following the

Younger Dryas, a trend discussed further in the following section.

141

Microblades, Climate, and Ecology

Paleobotanical studies from interior Alaska have suggested that the spread of woody species fueled human expansion throughout eastern Beringia during the Younger Dryas

(Bigelow 1997:168). Microblade production at Dry Creek component II in the Nenana

Valley may have been a response to the onset of the Younger Dryas cold event (Hoffecker and Elias 2003:44). Powers and Hoffecker (1989:284) suggested the early Holocene transition to microblade production around 12.4 ka (10,500 14C BP) in the Nenana valley is a reflection of “more intensive exploitation of local resources” by increasingly variable subsistence strategies and toolkits in response to the post-Carlo advance associated with the Younger Dryas. This analysis supports that observation, but it must be noted that non- microblade components outnumber those with microblades during the middle and late

Younger Dryas. One interpretation of this could be a response to increased aridity during the Younger Dryas period (Abbott et al. 2000; Bigelow and Edwards 2001; Peteet and

Mann 1994) and lower than normal snow fall. However, it is more likely that this trend is primarily the result of the four separate loci at the Mesa site, each with multiple hearth features dated between 12 and 10.5 ka (see Table 7.9 and references within) which probably artificially inflated the number of non-microblade occupations represented in this analysis during and immediately after the Younger Dryas. Whatever the reason, the results of the current methodology failed to meet my initial expectation that microblade production should dominate components dating to the Younger Dryas period based on the assumption that it was a cold adapted technology.

Occupations lacking microblades but containing bifacial components have also been dated to the MTM period at Houdini Creek (Bowers et al. 1995), Jay Creek Ridge 142

(Dixon 1999; Dixon et al. 1985) and Susitna River Overlook (Wygal and Goebel 2006).

Superficially, these sites suggest a reduction in the use of microblades in central Alaska during unusually warm periods, but it is clear that no significant reduction took place because many components containing microblades occur during warm intervals (Fig. 5.1 and 5.2). However, microblade-bearing components at CZ 1 and CZ 2 from Broken

Mammoth may correlate with middle Holocene cooling trends. The CZ 2 occupation dates to the younger-Younger Dryas or Mesoglacial and CZ 1 dates between 4.5 and 4.6 ka

(Krasinski 2005:29; Yesner et al. 1992; Yesner and Pearson 2002:145), consistent with one of the mid-Holocene neoglacial events documented at Dry Creek (Powers and Hoffecker

1989:267, 276).

The most significant observation revealed here is that microblade industries were critical for arctic foragers as they struggled to survive the rapid onset of cold episodes during the last 14 ka in eastern Beringia. Microblade use increased over time even though overall occupation events declined after 11 ka. The pattern is consistent with the decreasing productivity of the mammoth steppe and spread of the northern forest. Whether or not cold climates prolonged seasonal winter conditions and increased the necessity for microblade use was difficult to demonstrate in this analysis, primarily because of problems associated with precisely linking human occupations to specific climatic events. However, the attempt was interesting in that it indicated major climatic shifts probably had a significant impact on the relative number of people in interior eastern Beringia at a given time. In the next section, I attempt an alternate approach at predicting the presence or absence of microblades in particular sites on the landscape with a seasonal toolkit hypothesis based on ethnographic accounts and specialty site types. 143

Seasonality, Landuse, and Toolkit Variability

Gal (2002) proposed a seasonal model for addressing variability in bifaces and microblades in the Brooks Range and Rasic (2008) expanded the concept convincingly to explain lithic variability among early assemblages in northwestern Alaska. In 2008, I presented an initial attempt to apply a similar seasonal model to sites in central and southcentral Alaska (Wygal 2008a). Esdale (2009) developed the approach further to explain Northern Archaic assemblage variability and attributed microblade core style to seasonality as opposed to the cultural historical approach which inherently associates tool type with cultural preference or ethnic affiliation. In this section, I have continued with this line of inquiry in southcentral Alaska by integrating ethnographic reports from the middle Susitna River Dena’ina seasonal subsistence rounds (Kari and Fall 2003) and the known prehistoric use of ice patch kill sites in alpine settings during the fall season

(VanderHoek et al. 2007a, VanderHoek et al. 2007b; Wygal 2008b).

The Gal Seasonality Model. Gal (2002) interpreted variability among lithic scatters in the Brooks Range as the product of a seasonal approach to technological organization.

He proposed that microblade cores were mass produced in the autumn in preparation for winter when toolstone was difficult to acquire. To ensure sufficient quantities of toolstone, high quality cores were curated to last through the winter months. Microblade core production sites contain relatively high numbers of microblade fragments. Discarded microblade cores and rejuvenation tablets are also abundant as striking platforms and channel facets were being prepared, tested, and discarded if flawed. During the winter, Gal

(2002) predicted that microblade production and utilization would be exceedingly frugal with blades detached only when necessary, such as when replacement inset segments were 144

needed. Thus, only the occasional exhausted core and relatively few microblade fragments are expected in winter occupations. During the summer, bifaces were favored over microblade technology as toolstone was in relatively ample supply.

This model fits well with Elston and Brantingham’s (2002) argument that biface armatures are more lethal but an inefficient use of toolstone compared to insetting microblades in osseous points. The osseous component is less brittle and therefore more effective in winter months when compared to biface armatures; however, the amount of time required to construct a microblade composite projectile is far greater than a bifacial one (Elston and Brantingham 2002). For these reasons, winter and summer toolkits are expected to be different.

Ethnoarcheological studies of the Nunamuit of Anaktuvuk Pass indicated seasonal adjustment between subsistence strategies and specific tasks. Preparation of task-specific toolkits was undertaken accordingly and prior to setting out from a residential camp

(Binford 1978a, 1978b, 1980). Upper Cook Inlet Dena’ina also practiced seasonal subsistence strategies and often cached resources on the landscape for future seasonal rounds (Kari and Fall 2003). In the summer season, salmon are harvested, processed and stored by the quesqa or big man in specialized facilities. Berries, sheep, caribou, moose and bear are taken in the fall. Although small game and furbearers are the primary game species during winter, bear and moose are also hunted. Most activities in the winter are directed toward the potlatch, a function made possible by the seasonal abundance provided by the salmon harvest and storage (Kari and Fall 2003), a resource not available to late

Pleistocene and early Holocene foragers (Bacon et al. 1983). During the early spring,

“trade and visiting” was undertaken and freshwater fish and waterfowl were consumed 145

(Kari and Fall 2003). The Dena’ina seasonal strategy is a fair representation of early

Alaskan foragers excluding the abundance of salmon runs because of high glacial silt loads in the rivers. Although moose and wapiti would have been lean and more difficult to hunt in winter, these species would have been taken opportunistically in the cold season. Bison were an easier target because they associate in herds.

Other relevant observations (Esdale 2009) include an association between bison remains and microblades from early Holocene sites in the Tanana Valley (Holmes and

Bacon 1982). Potter (2008a) correlated microblade technology with wapiti, moose, and bear, while bison, sheep and caribou were frequently associated with biface armatures in archeological sites from interior Alaska. This is significant if one accepts the seasonality model. Esdale (2009:368, citing Binford 1980) reported that the Nunamuit preferred stone points for summer prey “such as bear” but the species was not indicated. Brown bears

(Ursus arctos) are in the worst condition in June according to the Nunamuit, and they reported taking many brown bears as late as December (Rausch 1951:166-167, 170, 181).

It is possible they were referring to using bifaces on polar bears (Ursus maritimus) which would be more vulnerable during the summer when on land. In the Susitna valley,

Dena’ina ethnographies report bear hunting in the winter, using specialized dogs to locate dens (Kari and Fall 2003:23) thereby neutralizing danger by dispatching bears as they slept.

Minimizing risk to the hunter was an important factor in deciding which armature to employ (Bamforth and Bleed 1997; Elston and Brantingham 2002; Torrence 1989).

Biface projectiles were less reliable and failed more frequently than composite tools, and they were also difficult or impossible to repair, especially when hafted. However, bifaces 146

are more lethal and inflict greater physical damage by penetrating deeper into the prey

(Elston and Brantingham 2002).

Ellis (1997) has evaluated the factors influencing choice of stone or organic points in a survey of over 100 ethnographic cases. He found that stone arrow and spear points are used almost exclusively for large and dangerous game (bears and large herd animals) and in warfare where killing the enemy is the object. However, the use of thrusting spears with stone points tended to be limited to situations in which there was little danger to the user or in which a number of replacement weapons were available. Organic points were used when wielding thrusting spears against smaller herd animals, against dangerous animals when no replacement weapons were available, when multiple thrusts were to be given in warfare, on arrows used in cold weather (when stone points are brittle and easily broken), and when throwing spears in heavy underbrush (Elston and Brantingham 2002:104).

The observations made by Ellis (1997) and by Elston and Brantingham (2002) explain many aspects of microblade use. For example, it illustrates why Clovis hunters apparently abandoned microblade composite weapons from their repertoire. Not only was it found in southerly regions significantly warmer than northern latitudes, but Clovis also focused on large bodied and therefore more dangerous prey species (Waguespack and

Surovell 2003). The same pattern occurred in southern Europe where Solutrean foragers also hunted large bodied game but are not known for microblade technology (Bradley and

Stanford 2004).

In Alaska, lanceolate-style projectiles co-occur with microblades in many of the largest archeological components in the interior (Hamilton and Goebel 1999; Hoffecker in press; Powers and Hoffecker 1989). A microblade component was even found among the

Paleoindian style Mesa points at the Mesa and Putu sites, although these were interpreted as unrelated deposits by the original investigators (Bever 2006, 2008). Given these associations, it may no longer be tenable to attribute these two fundamentally different stone tool production trajectories as indications of different cultural affiliations or waves of migrations into or throughout Beringia. Instead, it seems the various culture groups living 147

in interior Alaska have used microblades and lanceolate biface technologies simultaneously and in similar ways based on functionality and risk minimizing strategies.

Annual Economic Model for Southcentral Alaska. Utilizing aspects of technological organization, landuse, and subsistence strategies, I developed a testable annual economic model depicted in Figure 5.3. This model is intended to predict the presence or absence of microblades in some sites, but it may not be applicable to all situations. Because human behavior can be unpredictable, this model is intended only to move the current archeological dialogue in a new direction, away from typology and culture history.

Fall would have been the most productive season for early foraging societies, as caribou and sheep could be easily dispatched in large numbers from predictable alpine ice patches. VanderHoek et al. (2005, 2007a, 2007b) reported that caribou and other herd animals are attracted to ice patches as a refuge from the heat and mosquitoes during the summer and fall seasons. Once a herd of caribou occupy an ice patch, they are reluctant to leave, making them an easy target for ambush. Although this behavior occurs throughout the heat of summer, it becomes concentrated in the fall when only permanent or year-round snow patches remained in highly predictable settings making fall the ideal time for sheep and caribou hunting.

Ice patch finds from southern Yukon (Hare et al. 2004) and Alaska (Dixon et al.

2005; VanderHoek et al. 2007a) primarily contain killing implements, like complete points, distal point fragments, snares and faunal elements. Presently, there is only one example of a microblade composite tool in an ice patch kill site. That specimen was recovered in the southern Yukon and dated to 8.2 ka (8110 ± 50 cal BP/7310 ± 40 BP) 148

(Helwig et al. 2008). The majority of finds are bifacial and I am unaware of published accounts of ice patches containing proximal point fragments or evidence of tool repair (i.e. burins, hammerstones, or cores), because these sites were the primary loci of direct procurement, not carcass processing or re-tooling. Because alpine (>914.4 m/3000 ft asl) sites were most likely uninhabitable during the winter, artifact types found in alpine settings were likely deposited in the summer or fall seasons (Wygal 2007b, 2008a). Given that biface manufacturing was the primary activity at Costello Creek (1038.8 m/3408 ft asl) and Bull River II (1030.8 m/3382 ft asl), I conclude these sites were occupied in the late summer or even the early fall and that biface, not microblade production, was undertaken in preparation for upland hunts (Chapter IV).

Following the fall hunt, preparation of winter toolkits ensued with the mass production of idealized microblade cores as described in the Gal (2002) Seasonality

Model. Representative of this behavior are Whitmore Ridge, Phipps (ca. 883.9 m/2900 ft asl), and Teklanika West (761.1 m/2497 ft) sites given the high frequencies of recovered microblades and cores, but these sites also contain finished bifaces, indicating they were probably transitional between summer and fall toolkits. After processing and caching game, and gearing up for winter, foragers would return to larger river valleys along the banks of the upper or middle Susitna River (or similar settings) where residential base camps may have been established and supported by task specific hunting sites. Late Fall to early winter was the time for trade and “visiting” among the upper Cook Inlet Dena’ina

(Kari and Fall 2003) and the same was likely true in the early Holocene. 149

Figure 5.3. The Annual Economic Model. Shaded areas depict seasons of relative abundance. Toolkit variability appears in the outer circle; the inner circle depicts landuse and subsistence strategies.

Late winter would have been more difficult as food and raw material stores ran low and groups would have to disband and fan out in search of game. At such times, people may have had to move more frequently between short-term hunting camps and conserve their raw materials. This probably occurred in a variety of settings including the forested lowlands, where game would be more easily targeted than in summer months because of snow cover and the absence of forest undergrowth. Evidence of this behavior may come in 150

the form of relatively small tools, frequent tool recycling, little to no cortex, and exhausted microblade cores such as end-style cores and cores made on thick flakes or facies of broken tools. Given the nature of the TCO assemblages, particularly component I, and its location in the Susitna River lowlands, the component is interpreted as a late winter to early spring toolkit refurbishing and overlook occupation.

After breakup when river bars were released from winter’s grip, a decreasing reliance on microblade composite tools would give way to uninhibited tool production and foragers would begin to migrate up the streams and tributaries of the Susitna and Chulitna

Rivers toward summer and eventually fall hunting grounds. Summer would have been unusually lean without annual salmon runs, as game is especially difficult to locate in the thick forested lowlands. But lakes, ponds and tributaries of montane settings would become increasingly occupied as people moved into the littoral zones between the lowlands and alpine settings. Waterfowl and fish would have been plentiful, and bugs are fewer at higher elevations. In the later summer months, middle alpine and montane settings are ideal for harvesting berries, which generally ripen in August just prior to the fall hunt.

Conclusions

A unifying interpretation on the late Pleistocene and early Holocene culture history of eastern Beringia has eluded Alaskan archeologists for more than 40 years, in no small part due to the dichotomy that exists between the appearance of microblades in some contexts but not in others. Exploring the multiple factors affecting the presence or absence of microblades among assemblages might be more productive than establishing a consensus in the cultural historic approach. In this study, radiocarbon data from more than 60 151

components in eastern Beringia were systematically evaluated for reliability. Dates deemed most representative of occupation events were compiled in 500 and 250 year intervals and compared to worldwide climatic conditions. The results suggest populations were high and microblade use was low during the Allerød and post-Younger Dryas warm trends. Cold episodes like the Older Dryas, younger-Younger Dryas, and two middle

Holocene neoglacial periods saw significant reductions in overall occupation events suggesting rough times for all foragers regardless of their toolkit during the onset of the coldest periods. Most significant of these was the population bottleneck coincident with the onset of the Younger Dryas; however, by the end of the Younger Dryas, population levels are at their highest and microblades appear to be scarce. Occupations also declined toward the end of the MTM, although it is uncertain if this trend was in response to increased aridity or the rise of the boreal forest. Admittedly, it is difficult to assess the distribution of data presented in Figures 5.1 and 5.2, in part because slight differences in the organization of data points could cause significantly different patterns in the distribution of microblade use over time. Nevertheless, microblade industries appear to predominate in lithic assemblages during or following population nadirs. In that sense, the data presented here are interpreted as a reflection of technological fitness among foragers using microblade industries during unusually distinct climatic and ecologic transitions.

A finer grained seasonal approach may be more successful in interpreting the presence or absence of microblades in archeological sites. Originally presented by Gal

(2002) and developed further to fit regionally specific datasets, the predictive model assumes seasonal toolkits were employed in response to task specific uses in response to toolstone limitations during winter months. The model incorporates ethnographic accounts 152

of seasonal subsistence activities and applies landuse strategies in a method that explains the microblade/non-microblade dichotomy in a wider economic and social system.

153

CHAPTER VI

HUNTER-GATHERER TECHNOLOGICAL ORGANIZATION

IN SOUTHCENTRAL ALASKA

Several key questions were raised in Chapter I regarding differences observed in the archeological record between bifacial and microblade technologies. How were technology and settlement affected by extreme environmental fluctuations between 14 and

6 ka? How can differences in technology and landuse be explained by the extremity of the seasonal cycle in the north? Which of these two factors, long term climate change or annual seasonality, had the greatest impact on prehistoric adaptations?

In the previous chapters, I argued that few small scale composite refurbishing sites occur in alpine settings because these areas were occupied in the late summer to early fall when bifaces were produced for specialized caribou and sheep hunting. I also reiterated the Elston and Brantingham (2002) argument that microblade technology was more effective than bifaces in cold conditions. Working from these assumptions, I developed the Annual Economic Model for tool use in central and southcentral Alaska, inspired by

Gal’s (2002) presentation on frugality in microblade use. Based on this premise, microblade use was coincident with the winter season when raw material was scarce, while bifacial armatures were utilized more frequently in the summer months when lithic sources were generally abundant. It can be difficult to test this hypothesis at large sites with palimpsest assemblages, like at Dry Creek, but in this dissertation I have used small sites, unaffected by extensive use and primarily dedicated to single or limited activities by a relatively low number of individuals to provide better data for when and where specific tool types were employed (sensu Hall 1982). 154

In subsequent chapters enough data were presented to accept the hypothesis that on a macro-scale there is some correlation between shifts in the predominance of microblade technology and paleoenvironmental change over time. I have also used current knowledge of adjoining regions in the central Alaska Range to help place the earliest sites known from the study areas into a model of human exploration and occupation of southcentral Alaska. Combined, these related themes inform us about the initial human settlement of southcentral Alaska and relate to other pertinent questions about the archeology of the north. In this final chapter, I integrate the data through a series of analytical intersite comparisons of artifact attributes related to lithic reduction to test the primary hypothesis that activities undertaken in the lowlands differed fundamentally from those in alpine settings. Finally, I use these results to develop a testable model for explaining toolkit variability in central and southcentral Alaska based on the presence or absence of microblade related technologies.

Intersite Statistical Analysis

In Chapters III and IV, I concluded that the TCO components resulted from the conservation of raw material during winter-time refurbishing activities. Activities at SRO were interpreted as the expedient production of heavy chopping and cutting implements related to large game butchering and marrow extraction. Although butchering sites could occur anywhere on the landscape, the nature of the behaviors reflected in the TCO-I and

TCO-II assemblages were arguably the result of seasonality, assuming the lowlands were more frequently occupied during the winter, when raw materials became difficult to obtain and microblade technologies were more effective. At the Bull River II and Costello Creek sites in the uplands of the central Alaska Range, initial core reduction and biface 155

manufacture constituted the primary activities; raw materials were easily available and microblades were not necessary because these sites were presumably occupied late in the summer season. These interpretations are summarized in Table 6.1.

Table 6.1. Summary of site settings, primary activities, lithic use, and season of occupation. Component Setting Type Lithic Use Season TCO-I lowland refurbishing conservative winter TCO-II lowland refurbishing/hide working conservative winter SRO lowland butchering liberal indeterminate BR upland primary reduction/biface production liberal summer CC upland primary reduction/biface production liberal summer

In testing the seasonality hypothesis, I have conducted a series of intersite statistical tests of equality for lithic attributes, primarily on flaked debris, aimed at identifying similarities in patterns and relationships in the activities undertaken between topographic settings. The tests varied according to the nature of the data, but because raw material use and core reduction strategies are at the heart of the Annual Economic Model presented in

Chapter V, many of the tests focused on raw material reduction and use.

The Levene’s two-sample T test and one-way ANOVA test of means were used for assessing similarities in reduction intensity between components and these were followed by a Tukey (1977) Post Hoc test to assess where differences and similarities occurred.

The Mann-Whitney test is a rank based non-parametric extension of ANOVA capable of testing the relationship between two independent samples and the test does not require that the data be equally distributed (Mann and Whitney 1947). Because two components were associated with biface production in the uplands and two refurbishing sites in the lowlands, the Mann-Whitney statistic was also useful in determining if mean frequencies of the degree of dorsal cortex and flake size classes (ordinal scale data) were equal between components in the same topographic setting. The non-parametric Kruskal-Wallis test of 156

means is an extension of the Mann-Whitney statistic (Kruskal and Wallis 1954) and was used to further compare lithic reduction strategies between three or more components using the degree of dorsal cortex and flake sizes between different occupation assemblages. To

further compare these results to lithic reduction strategies between lowland and upland

settings, a Pearson’s chi-square was used to test for equality in the distribution of mean

frequencies on sets of nominal data (i.e., debitage, raw material, and platform types).

Table 6.2 presents a crosstabulation of raw material types and flaked debris from each of the five archeological occupations. Siltstone and basalt are the most frequently utilized raw materials, and raw material diversity is highest in the refurbishing occupations

TCO-I and TCO-II. Although there are some similarities in the distribution of raw material types between SRO, TCO-I, and TCO-II, SRO is clearly dominated by basalt.

Table 6.2. Intersite crosstabulation of flaked debris by raw material Raw Material Type Component Chert Siltstone Chalcedony Claystone Basalt Sandstone Obsidian Schist Slate Total Count 129 6 2 — 75 4 4 28 4 252 TCO-I % of row 51.2 2.4 0.8 — 29.8 1.6 1.6 11.1 1.6 100 Count 87 14 5 — 125 5 45 22 5 308 TCO-II % of row 28.2 4.5 1.6 — 40.6 1.6 14.6 7.1 1.6 100 Count 17 26 — — 295 66 — — — 404 SRO % of row 4.2 6.4 — — 73.0 16.3 — — — 100 Count 16 501 2 4 5 38 — — — 566 BR % of row 2.8 88.5 0.4 0.7 0.9 6.7 — — — 100 Count 2 389 — — 9 22 — — — 422 CC % of row 0.5 92.2 — — 2.1 5.2 — — — 100 Count 251 936 9 4 509 135 49 50 9 1952 Total % of row 12.9 48.0 0.5 0.2 26.1 6.9 2.5 2.6 0.5 100

There are similarities between the variety of lithic types found in the upland sites

where siltstone predominates and other lithic types are less common (Fig. 6.1). In the

lowland assemblages, it has not been determined whether the remaining lithic varieties in

these sites are local or non-local in origin, save for the obsidian pieces traced to Batza 157

Téna. However, similar materials to those from the lowland archeological assemblages have been observed in various gravel bars and tributaries surrounding the sites, but these have yet to be chemically quantified (Coffman 2006). Thus, given the readily available assortment of raw materials in the Trapper Creek region, it is assumed that all of the

non-obsidian materials found at TCO and SRO were available within 30 km of the sites.

100.0% TCO-I 90.0% TCO-II 80.0% 70.0% SRO 60.0% Bull River II 50.0% 40.0% Costello Creek 30.0% 20.0% 10.0% 0.0%

Figure 6.1. Frequency of raw materials among flaked debris from all components.

One prominent indicator of raw material reduction is the degree of dorsal cortex on

flaked debris (Table 6.3). A Kruskal-Wallis test was used to compare the degree of dorsal

cortex on debitage between all five components and indicated significant differences did

occur (4 df, p = 0.00). Differences also occur when SRO is withheld from the analysis and the amount of dorsal cortex from only the refurbishing/lowland components at TCO are compared to the biface manufacturing/upland components at Bull River II and Costello (3 df, p = 0.000). There were also differences between SRO and the upland sites (2 df, p =

0.000). The degree of dorsal cortex is equivalent among all three components in the

Susitna River lowlands (2 df, p = 0.093). Because some of the assemblages had relatively

low numbers of artifacts with cortex, I collapsed the variables to reflect only the presence 158

or absence of cortex on flaked debris and repeated the Kruskal Wallis tests with similar

results (see bracketed statistics in Table 6.3).

Table 6.3. Intersite crosstabulation of flaked debris by degree of dorsal cortex Dorsal Cortex* Component 0% 1-10% 10-50% 50-90% >90% Total TCO-I Count 234 7 2 3 6 252 % of Row 92.9 2.8 0.8 1.2 2.4 100 TCO-II Count 269 11 7 6 15 308 % of Row 87.3 3.6 2.3 1.9 4.9 100 SRO Count 363 8 14 8 11 404 % of Row 89.9 2.0 3.5 2.0 2.7 100 BR Count 429 6 21 12 98 566 % of Row 75.8 1.1 3.7 2.1 17.3 100 CC Count 355 4 8 6 49 422 % of Row 84.1 0.9 1.9 1.4 11.6 100 Total Count 1650 36 52 35 179 1952 % of Row 84.5 1.8 2.7 1.8 9.2 100 * Tests on presence/absence of cortex appear in brackets below (see text for details). Kruskal-Wallis results on dorsal cortex between the following components: All components: X2= 64.234, 4 df, p = 0.00 [X2 = 57.013, 4 df, p = 0.000] SRO, BR and CC: X2 = 37.506, 2 df, p = 0.00 [X2 = 33.333, 2 df, p = 0.000] SRO, TCO-I, and TCO-II: X2 = 4.753, 2 df, p = 0.093 [X2 = 4.619, 2 df, p = 0.099] TCO-I, TCO-II, BR, CC: X2 = 47.227, 3 df, p = 0.000 [X2 = 42.892, 3 df, p = 0.000]

These results are expected given the relatively high frequency of flakes with more

than 90% dorsal cortex in the upland sites versus the lowland occupations which have

relatively low frequencies of cortex on flakes. Among the lowland sites, none have more

than 5% of the flaked debris with > 90% dorsal cortex and less than 2% of these

assemblages have 50-90% dorsal cortex. These frequencies indicate that early stages of

lithic reduction were undertaken in the uplands, while later stages related to more refined

retouching activities were more common in lowlands.

To further examine reduction strategies, a series of Kruskal-Wallis tests were applied to test for equality of means of flake size classes between all five components, and then repeated in the same fashion described above. Results rejected the null hypothesis under all scenarios, each producing P values of 0.00, first when all components were 159

compared, then in comparison of lowland butchering (SRO) and lowland refurbishing

(TCO-I/TCO-II) occupations, lowland butchering (SRO) with upland biface production

sites (BR/CC), and finally lowland refurbishing (TCO-I/TCO-II) with upland biface production occupations (BR/CC). None of the components had an equal distribution of mean flake size classes (Table 6.4).

Table 6.4. Intersite crosstabulation of flaked debris by size class Debitage Size Class (cm)* Component < 1 1-3 3-5 5-7 7-9 9-11 11-13 13-15 Total TCO-I Count 72 142 27 10 1 — — — 252 % of row 28.6 56.3 10.7 4.0 0.4 — — — 100 TCO-II Count 105 159 33 9 1 — — 1 308 % of row 34.1 51.6 10.7 2.9 0.3 — — 0.3 100 SRO Count 44 157 96 57 28 17 4 1 404 % of row 10.9 38.9 23.8 14.1 6.9 4.2 1.0 0.2 100 BR-II Count 17 323 160 52 9 5 — — 566 % of row 3.0 57.1 28.3 9.2 1.6 0.9 — — 100 CC Count 7 271 106 26 9 2 1 — 422 % of row 1.7 64.2 25.1 6.2 2.1 0.5 0.2 — 100 Total Count 245 1052 422 154 48 24 5 2 1952 % of row 12.6 53.9 21.6 7.9 2.5 1.2 0.3 0.1 100 * Classes >9 cm were collapsed into a single variable in the following analyses. Kruskal-Wallis on flake size class between the following components: All components: X2= 263.227, 4 df, p = 0.00 SRO, BR, and CC: X2 = 18.216, 2 df, p = 0.000 SRO, TCO-I, and TCO-II: X2 = 158.479, 2 df, p = 0.000 TCO-I, TCO-II, BR, CC: X2 = 225.287, 3 df, p = 0.000

The above tests imply significant differences in how stone was worked between

different topographic settings, but they do not explicitly detail where these differences

occur. To better understand the lithic reduction activities, the Mann-Whitney test was

used to test for equality between two individual components. In comparing the TCO-I and

TCO-II refurbishing lowland occupations, the degree of dorsal cortex continues to be

significantly different, but mean flake sizes are considered equivalent (Table 6.5). The

same results occur when comparing Bull River II and Costello Creek; flake size classes are

equal but the degree of dorsal cortex is different (Table 6.6). 160

Table 6.5. Mann-Whitney tests on TCO-I and TCO-II degree of cortex and flake size among debitage Ranks Component N Mean Rank Degree of TCO-I 252 271.88 (272.00) Dorsal Cortex TCO-II 308 287.56 (287.45) Total 560 Size Class TCO-I 252 288.70 TCO-II 308 273.79 Total 560 Results Cortex Cortex* Size Class Mann-Whitney U 36634.500 36666.000 36742.000 Wilcoxon W 68512.500 68544.00 84328.000 Z -2.175 -2.147 -1.204 Asymp. Sig. (2-tailed) 0.030 0.032 0.229 * Separate test on presence/absence of cortex only, mean ranks in parentheses Grouping Variable: TCO-I and TCO-II

Table 6.6. Mann-Whitney tests on Bull River II and Costello Creek degree of cortex and flake size among debitage Ranks Component N Mean Rank Degree of Bull River II 566 511.96 (512.07) Dorsal Cortex Costello Creek 422 471.08 (470.08) Total 988 Size Class Bull River II 566 505.27 Costello Creek 422 480.06 Total 988 Results Cortex Cortex* Size Class Mann-Whitney U 109544.500 109480.000 113330.50 Wilcoxon W 198797.500 198733.000 202583.500 Z -3.159 -3.197 -1.573 Asymp. Sig. (2-tailed) 0.002 0.001 0.116 * Separate test on presence/absence of cortex only, mean ranks in parentheses. Grouping Variable: BR and CC

In comparing assemblages within the same topographic setting, the degree of dorsal cortex fails to meet the expectations of equality, but flake size classes are equivalent within the same location. Results from the flake size class analyses indicate that lithic reduction behaviors were undertaken in a similar pattern during refurbishing activities in the Susitna

River lowlands, and the same is true for biface production in the uplands near Broad Pass. 161

However, there appears to be significant intra-regional variability in the degree of dorsal

cortex primarily because of differences in how lithics were being reduced in these

locations. Because the results from tests on the degree of dorsal cortex failed to meet my

expectations, testing the hypothesis on different attributes is necessary.

Reduction stage and intensity were further examined by categorizing each

assemblage based on debitage weights. The weights of individual debitage pieces are an exact measure of flake mass and thus an indication of reduction stage because smaller mean weights of flaked debris represent smaller flake sizes (Odell 2004:126; Shott

1994:80). Smaller flake weights should occur in later stages of the reduction cycle; for example, tool resharpening and refurbishing activities generate smaller detritus than do initial core reduction or bifacial blank production. In a sense it is also an indication of reduction intensity because less detritus is removed during refurbishing than in tool production. In times of raw material stress, higher frequencies of retouch chips and small bifacial thinning flakes should be more common as tool recycling, resharpening, and repair take precedence over new biface production. These activities are reflected in smaller average flake weights, size classes, and associated debitage types. Among the five components, individual debitage weights are heaviest at SRO, followed by the upland sites.

The refurbishing occupations at TCO have the smallest average debitage weights.

To test the hypothesis that reduction intensity at Bull River II and Costello Creek were equivalent, Levene’s two-sample T test for equality of variance was used to test if the mean weights (g) of debitage were statistically equivalent. The results failed to reject the null hypothesis indicating the mean weights of flaked debris at Bull River II and Costello

Creek are statistically equivalent (t = -0.231, p = 0.817). The same test was applied to 162

mean debitage weights (g) at TCO-I and TCO-II. Like the previous test, this test failed to reject the null hypothesis indicating individual weights of flaked debris are statistically equivalent (albeit barely) at TCO-I and TCO-II (t = -0.659, p = 0.51).

While no significant differences occur in mean debitage weights between TCO I

and II, and Bull River II and Costello Creek, I additionally utilized a one way ANOVA calculation to test the null hypothesis that mean debitage weights between all five occupations, including SRO, were statistically equivalent. The ANOVA test

demonstrated that there was a significant difference in the mean weights of flaked debris

between these components (F = 18.156, 4 df, p = 0.000). A Tukey Post Hoc test was then employed to determine where the differences occurred (Table 6.7). The results of these tests confirmed the T tests and indicated the mean debitage weights from the TCO-I and

TCO-II assemblages are statistically equivalent, as are the Bull River II and Costello Creek assemblages, but these two groups are distinct from one another. The SRO assemblage was statistically different from the four other sites in the analysis (Table 6.8), implying that there are significant differences in mean debitage weights between the upland sites, lowland refurbishing sites, and the butchering occupation at SRO. These differences are the result of differential reduction strategies.

The average weight of the debitage at SRO (12.6 g) was nearly twice those at Bull

River II (6.2 g) and Costello Creek (6.6 g) and this is consistent with the overall size of the artifacts produced at SRO but also in the lack of secondary and tertiary flaking at SRO.

Although primary core reduction was prevalent in the upland sites, there was also a degree of secondary flaking as the bifaces were formed. The average weights of debitage at

TCO-I (1.97 g) and TCO-II (2.4 g) is less than half of the mean weights found at Bull River 163

II and Costello Creek. This is consistent with the conservation of lithic resources through simple tool resharpening and refurbishing activities instead of tool production.

Table 6.7 Flaked debris Tukey post hoc test of weight means. 95% Confidence Interval Mean Difference Lower Upper (I) (J) (I-J) Std. Error Sig. Bound Bound TCO-I TCO-II 0.42734 1.59230 0.999 -3.9202 4.7748 SRO -10.18140* 1.50476 0.000 -14.2899 -6.0729 BR -3.87554 1.41962 0.050 -7.7516 0.0005 CC -4.18347* 1.49238 0.041 -8.2582 -0.1088 TCO-II TCO-I -.42734 1.59230 0.999 -4.7748 3.9202 SRO -10.60875* 1.41801 0.000 -14.4804 -6.7371 BR -4.30288* 1.32733 0.011 -7.9269 -0.6788 CC -4.61082* 1.40487 0.009 -8.4466 -0.7751 SRO TCO-I 10.18140* 1.50476 0.000 6.0729 14.2899 TCO-II 10.60875* 1.41801 0.000 6.7371 14.4804 BR 6.30587* 1.22093 0.000 2.9723 9.6394 CC 5.99793* 1.30481 0.000 2.4354 9.5605 BR TCO-I 3.87554 1.41962 0.050 -0.0005 7.7516 TCO-II 4.30288* 1.32733 0.011 0.6788 7.9269 SRO -6.30587* 1.22093 0.000 -9.6394 -2.9723 CC -0.30794 1.20564 0.999 -3.5997 2.9839 CC TCO-I 4.18347* 1.49238 0.041 0.1088 8.2582 TCO-II 4.61082* 1.40487 0.009 0.7751 8.4466 SRO -5.99793* 1.30481 0.000 -9.5605 -2.4354 BR 0.30794 1.20564 0.999 -2.9839 3.5997 * The mean difference is significant at the 0.05 level.

Table 6.8. Tukey homogenous subsets of debitage weights Subset for alpha = 0.05 Component N 1 2 3 TCO-II 308 1.9666 TCO-I 252 2.3940 BR 566 6.2695 CC 422 6.5774 SRO 404 12.5754 Sig. 0.998 0.999 1.000 Means for groups in homogeneous subsets are displayed.

164

The statistical patterns revealed thus far distinguish upland biface production

facilities from the lowland refurbishing occupations and there is some evidence to suggest

the butchering activities at SRO are unique. A finer-grained interpretation of reduction

strategies can be assessed through additional intersite comparisons of debitage types and

raw material selection. Because these tests aim at determining equality of knapping

behaviors, further tests on platform preparation are included.

Because platform types are indicative of patterns in lithic reduction, e.g.,

percussion versus pressure flaking techniques; significant differences between the

frequencies of different platform types imply distinctions in flaking behaviors (Andrefsky

1998; Odell 2004). Chi-square was applied to test the hypothesis that the frequencies of

specific platform types on debitage pieces were distributed equally between the two upland

sites and the two refurbishing components at TCO. The SRO debitage assemblage was

withheld from these analyses because I was only interested in testing between refurbishing

and biface production activities in different topographic settings. It is also important to

note that debitage pieces that lacked platforms were also removed from the analyses and,

because of the low sample sizes, crushed platform types were withheld as well leaving a

total sample size of 964 lithic pieces with cortical, simple, and complex platforms among

the four components. The test rejected the hypothesis indicating a significant difference

between the use of different platforms types between the components (6 df, p = 0.000).

To better define these differences, a chi-square was applied to test the hypothesis

that the frequencies of platform types on debitage pieces were distributed equally only

between the two upland sites (2 df, p = 0.327) as well as between the lowland refurbishing sites (2 df, p = 0.293). These results failed to reject the null hypothesis indicating there 165

was no significant difference in platform preparation between TCO-I and TCO-II. There

were also no differences in the frequency of platform types between Bull River II and

Costello Creek (Table 6.9).

Table 6.9. Intersite crosstabulation of flaked debris by platform preparation. Platform Preparation* Component Cortical Simple Complex Crushed Absent Total Count 9 79 66 3 95 252 TCO-I % of row 3.6 31.3 26.2 1.2 37.7 100 Count 18 117 72 8 93 308 TCO-II % of row 5.8 38 23.4 2.6 30.2 100 Count 30 116 74 1 183 404 SRO % of row 7.4 28.7 18.3 0.2 45.3 100 Count 40 209 82 — 235 566 BR % of row 7.1 36.9 14.5 — 41.5 100 Count 43 158 71 — 150 422 CC % of row 10.2 37.4 16.8 — 35.5 100 Count 140 679 365 12 756 1952 Total % of row 7.2 34.8 18.7 0.6 38.7 100 * Tests run only on cortical, simple, and complex platforms. Pearson’s Chi-Square Test Results on: TCO-I, TCO-II, BR, CC: X2 = 27.350, 6 df, p = 0.000 (n = 964)1. TCO-I and TCO-II: X2 = 2.910, 2 df, p = 0.293 (n = 361)2 BR and CC: X2 = 2.235, 2 df, p = 0.327 (n = 603)3

These results are consistent with previous tests run on the mean frequencies of flake

sizes indicating similarities between lithic reduction activities in the lowland components

at TCO, and, based on the overall small size classes and lack of dorsal cortex in these components, flaking behavior can be characterized as cautious in nature. Similarities also exist among the upland assemblages where flake size classes and the degree of dorsal cortex were larger than in the lowland components because, in the uplands, raw material conservation and tool repair were not necessary. There was an ample supply of raw

1 0% of cells have expected count less than 5. The minimum expected count is 17.57. 2 0% of cells have expected count less than 5. The minimum expected count is 11.52. 3 0% of cells have expected count less than 5. The minimum expected count is 37.44. 166

materials in the nearby creeks during summer months and this resulted in the early stage production of bifacial implements and a reduction in flake and tool recycling.

A crosstabulation of debitage and cores by occupation demonstrates an array of flaking detritus between the assemblages (Table 6.10). The most frequent types of debitage in all components are flakes and flake fragments; however, it is significant to note that percentages of these are lowest in the refurbishing occupations. Relatively high percentages of complete and fragmented flakes are consistent with occupations where primary and secondary reduction activities prevailed. The frequencies of bifacial thinning flakes are similar across all components, but this is expected since bifaces were recovered in all of the assemblages except Costello Creek. Retouch chips and chip fragments occur most frequently in the refurbishing sites and comprise relatively low frequencies in the biface production sites. Microblades and related debitage are only found in the lowland refurbishing occupations while microblades and microblade cores, technical spalls and all detritus related to microblade production are lacking in the upland components as well as at

SRO (Fig. 6.2).

More than a third of debitage pieces have simple platforms and nearly two-thirds of debitage pieces are complete flakes or flake fragments implying initial and secondary core reduction were the predominant activities in both of the upland assemblages.

Additionally, high frequencies of complex platforms and bifacial thinning flakes relative to retouch chips indicate biface manufacture was also a common activity. Because these trends are not observed in the lowland occupations, it can be assumed that different flint knapping behaviors were undertaken.

167

Table 6.10. Intersite crosstabulation of cores and debitage types.

Component

Debitage Type TCO-I TCO-II SRO BR CC Total Flake fragment Count 68 62 118 201 140 589 % of row 11.5 10.5 20.0 34.1 23.8 100 Count 25 40 127 165 141 498 Complete flake % of row 5 8 25.5 33.1 28.3 100 Count 1 2 — — — 3 Bipolar flake % of row 33.3 66.7 — — — 100 Count 2 2 1 2 2 9 Blade-like-flake % of row 22.2 22.2 11.1 22.2 22.2 100 Count 10 13 — — — 23 Microblade % of row 43.5 56.5 — — — 100 Count 6 11 11 94 46 168 Primary cortical spall % of row 3.6 6.5 6.5 56 27.4 100 Count 4 14 21 25 13 77 Secondary cortical spall % of row 5.2 18.2 27.3 32.5 16.9 100 Count 5 17 8 2 1 33 Retouch chip fragment % of row 15.2 51.5 24.2 6.1 3 100 Count 42 66 30 6 3 147 Retouch chip % of row 28.6 44.9 20.4 4.1 2 100 Count 52 56 62 42 56 268 Biface thinning flake % of row 19.4 20.9 23.1 15.7 20.9 100 Count 3 2 — — — 5 Burin spall % of row 60 40 — — — 100 Count 25 13 25 24 16 103 Angular shatter % of row 24.3 12.6 24.3 23.3 15.5 100 Count 0 4 1 — — 5 Cobble fragment % of row 0 80 20 — — 100 Count 2 2 — — — 4 Split cobble % of row 50 50 — — — 100 Count 1 1 — — 1 3 Worked chert cobble % of row 33.3 33.3 — — 33.3 100 Count 2 3 — 5 3 13 Core fragment % of row 15.4 23.1 — 38.5 23.1 100 Count 2 — — — — 2 Microblade core % of row 100 — — — — 100 Count 2 — — — — 2 Multidirectional flake core % of row 100 — — — — 100 Count 252 308 404 566 422 1952 Total % of row 12.9 15.8 20.7 29 21.6 100

168

40.0%

35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0%

TCO-I TCO-II SRO BR CC

Figure 6.2. Frequency of debitage types among flaked debris from all components. Note that retouch chips and retouch chip fragments were lumped into a single category.

In the lowland assemblages, 60% of the debitage pieces at SRO are complete and fragmented flakes; bifacial thinning flakes comprise 15.3% and retouch chips less than

10% of the SRO assemblage. Nearly 30% of debitage pieces at SRO have simple platforms and more than 7% are cortical. Significantly, one third of tools at SRO have cortical platforms. At TCO-I and TCO-II, just over one third of debitage pieces are complete or flake fragments, respectively, and retouch chips occur in higher frequencies than the other components comprising ca. 18% of the TCO-I and 27% of the TCO-II assemblages. Bifacial thinning flakes are also prevalent, comprising 20.6% of the TCO-I 169

and 18.2% of the TCO-II assemblages. Platforms at TCO-I are mostly simple followed

closely by complex styles. The high number of retouch chips and bifacial thinning flakes, based on their relatively small size classes supports the conclusion that at TCO-I and

TCO-II greater care was taken to repair and reuse already existing tools rather than produce

new ones.

Based on all of these analyses, there are enough similarities between the TCO-I and

TCO-II components to conclude that, although separated by more than 700 years, the

activities undertaken at these two occupations were similar. The same cannot be said of

SRO where butchering activities resulted in different flint knapping behaviors.

Similarities also occur between the two upland occupations, where core reduction and

biface manufacturing occurred, but these activities differed from those at TCO and SRO.

However, it could be argued that the differences observed between the upland/production

and lowland/refurbishing occupations could be the sole function of site type and not a reflection of topographic setting. It is clear these patterns are a function of occupational activities but the question remains: Did these activities differ because of topographic/ecological settings? My argument has been based on the observation that there is a dearth of upland microblade refurbishing and lowland bifacial production facilities in the archeological record. Another clue has to do with raw material selection use between the regions. Raw material sources in the uplands are plentiful but toolstone selection is not particularly diverse. In contrast, the middle Susitna valley has a plethora of various toolstones in the various streams and gravel bars that crisscross the region, many of which are highly flakable, fine grained cherts, basalts, and chalcedonies. Why was it necessary to conserve toolstone over multiple occupations in the Susitna lowlands? Why 170

invest the time and energy into producing microblade composite tools, especially if they cause less physical damage than do biface armatures (Elston and Brantingham 2002) and if during summer months sizable cobbles of quality toolstone can be located anywhere there is a stream or riverbank, which is essentially everywhere in southcentral Alaska?

Answers to these questions are addressed in the following section.

The Prehistoric Colonization of Southcentral Alaska: Human Adaptations in a Post Glacial World

The foraging societies that colonized the Susitna River lowlands carried microblade composite weapons and occupied geological features not unlike their progenitors north of the central Alaska Range. The first verifiable trace of human activity in the lowland areas is dated sometime between 10 and 9.2 ka at the TCO and SRO sites. Based on the ephemeral nature of the cultural deposits and the detritus they left behind these groups operated in relatively small but specialized hunting parties (perhaps as few as three to four individuals). These interpretations are supported by the nearby sites Eroadaway, Carlo

Creek, and Panguingue Creek, which also contain relatively small clusters of task specific artifacts likely deposited by relatively few individuals engaged in hunting or game processing activities. Because of the topographic setting and small dispersed nature of these occupations, the colonization of the Susitna River lowlands probably began as limited forays into winter ranges by individual bands spreading out from traditional territories located north of the central Alaska Range.

Bull River II and Costello Creek are located in the mountains just south of Broad

Pass in Denali National Park and Preserve and are well positioned beneath a series of peaks that probably contained high numbers of caribou and Dall sheep during the late summer 171

months. Hunters quarried the lithic sources in these alpine zones to produce bifacial

weapons for use in the uplands where they could capitalize on the seasonal concentrations of animal resources in preparation for upcoming winter months.

Many of the late Pleistocene/early Holocene sites in the Nenana and Tanana River valleys occupy bluff top overlook positions which Guthrie (1983) argued were logistical spike camps. However, the Healy Lake site diverges from this common pattern because it reportedly contained artifacts reminiscent of more than one lithic industry and multiple activities. Thus, Healy Lake was originally considered a substantial base camp rather than a hunting outpost (Cook 1996; Yesner 2001:319). Whitmore Ridge and Butte Lake, in southcentral Alaska, are reminiscent of Healy Lake in that they contained both bifaces and extensive microblade production facilities, and are represented by relatively high artifact frequencies near large lakes (Betts 1987; West et al. 1996c). Dry Creek component II arguably fits this mold; although not along a lake shore, it is near Eight Mile Lake and adjacent to an excellent raw material source in the Dry Creek river bed. Activities undertaken at the site include butchering, microblade core preparation, and there are bifacial projectile points within the assemblage (Hoffecker et al. 1996; Powers and

Hoffecker 1989; Powers et al. 1983).

These types of sites probably represent late fall to early winter occupations, where remnants of late fall bifacial toolkits merge with activities associated with winter-gear up activities, i.e., intensive microblade core preparation activities. Late fall to early winter would have been a time of plenty when hunter-gatherers could congregate, trade, and visit prior to the lean times of late winter and early spring when groups reduced into smaller family units and hunted the river lowlands to survive. A similar pattern has been observed 172

among traditional Athapaskan groups in central and southcentral Alaska (Kari and Fall

2003). Although Athaspaskan economies relied heavily on the seasonal abundance of

salmon (a resource that likely did not materialize until the middle Holocene) and less on

big game hunting from overlook positions, it is reasonable to assume that a similar seasonal

round was practiced in the late Pleistocene and first half of the Holocene.

Figure 6.3. Annual Economic Model for southcentral and central Alaska (12.5 to 6 ka). Sites discussed in this dissertation are abbreviated above and referenced in previous chapters.

Figure 6.3 depicts the hypothetical pre-salmon seasonal rounds in more detail.

Areas occupied during late summer and early fall are shaded in orange and expected to 173

contain low frequencies of artifacts related to microblade composite tools and high frequencies of ice patches, bifacial projectile point manufacturing, and meat caching facilities. Blue shaded areas would have been occupied during the late winter when food and raw materials were in relatively short supply. Because raw materials were conserved during this time of year and inset composites were the weapon of choice, artifact assemblages are expected to contain burins, burin spalls, and some but very few discarded microblades and microblade cores. In fact, lithic sources may have been so heavily conserved in the late winter season that archeological traces of these activities are difficult to locate today. Grey regions were occupied during late fall and early winter. Late fall occupations are expected to contain the highest numbers of microblades and microblade cores as groups geared up for the upcoming winter season. They are also suspected to have been the most heavily occupied sites as bands congregated in larger numbers during periods of seasonal abundance. If this was the case, it would be expected that late fall occupations would contain the most lithic diversity as groups converged from different parts of the region and engaged in trade.

Seasonality cannot account for all of the variability observed in the archeological record. Ecological and environmental factors have long been recognized as the driving force influencing hunter-gatherer adaptations (Harris 1968; Kelly 1995; Steward 1972;

White 1959). Several post-glacial climatic shifts directly altered prehistoric lifeways in central Alaska and probably forced groups south of the central Alaska Range as they fled deteriorating conditions in the interior (Yesner 1998). Although details associated with world-wide proxy records are not yet completely understood in eastern Beringia, there is some evidence that many of the most significant climate shifts impacted the landscapes of 174

southern and central Alaska (Bigelow 1997; Bigelow and Edwards 2001; Edwards and

Barker 1994; Edwards et al. 2001; Elias 2001; Hoffecker and Elias 2003; Mann and

Hamilton 1995; Mann et al. 1998; Mann et al. 2001; Mann et al. 2002; Mason et al. 2001).

The Younger Dryas cold snap, dated between 12.7 and 11.5 ka (Brubaker et al.

2001; Elias 2001:10), was perhaps the most severe post-glacial climatic oscillation for

hunter gatherer societies in eastern Beringia. The abrupt severity of the Younger Dryas

coincided with the dissipation of the mammoth steppe (Guthrie 1990, 2001, 2006), and the

cumulative results of these events were concentrated between 12.9 and 12.7 ka when

human populations in eastern Beringia reached levels so low that traces of their

occupations are not visible archeologically. It is clear that microblade technology was

crucial to groups that survived this population nadir because all of the occupations dated

within the 250-year interval following the crash are associated with microblade technology

(Chapter V, Fig. 5.2; Wygal 2007a, 2007b, 2008a, 2009a).

Implications for Major Theoretical Debates and Further Research in Eastern Beringia

Knowledge of the post-glacial colonization of southcentral Alaska hinges on a solid

understanding of the culture history in interior Alaska as well as the timing of deglaciation

in the middle Susitna River lowlands. On the latter point, there has been recent progress

in interpreting the glacial chronology of the central Alaska Range and upper Cook Inlet

areas (Briner and Kaufman 2008; Dortch 2006; Reger et al. 2007). However, these

findings do not necessarily inform about the timing of deglaciation in the specific areas

considered in this dissertation. Careful analysis of the loess deposits from the Trapper

Creek and Susitna River Overlook sites suggest deglaciation may have occurred earlier 175

than previously thought in the middle Susitna River valley, perhaps beginning immediately following the LGM, and the series of late glacial ice re-advances in the mountain valleys probably did not travel as far as Trapper Creek (Chapters II and III).

On the point of human colonization, I have argued that the first groups to colonize southcentral Alaska probably originated among foragers north of the Alaska Range, groups that in some circumstances employed microblade composite technologies. But no group carries their entire repertoire of tools with them at all times (Odess and Rasic 2007), so we should not assume that all components lacking microblades were produced by ethnically different people. It is not unreasonable to assume that under different circumstances these

groups switched between bifacial projectile points and microblade industries.

In many ways these issues directly relate to the long standing debate over the

Nenana and Denali complexes in our interpretation of the early prehistory of eastern

Beringia (Bever 2001b, 2006; Cook 1996; Goebel et al. 1991; Holmes 1998, 2001, 2009,

Hoffecker in press; Powers and Hoffecker 1989; Yesner 1996, 2001). In this dissertation,

I have purposely avoided addressing this issue primarily because the sites in southcentral

Alaska are typically associated with the Denali complex and there is little evidence to suggest the Nenana complex spread south of the Alaska Range (Yesner 2001).

Nonetheless, an informed debate over the source of the first foraging societies to live full time in southcentral Alaska cannot occur without addressing this ongoing discussion.

Was the Nenana complex related to the Denali complex and, thus, somehow associated with the colonization of southcentral Alaska? Given the stratigraphic separation of Nenana and Denali components in the Nenana valley, the original proposal by Powers and Hoffecker (1989) for two separate technological complexes in interior 176

Alaska during the terminal Pleistocene remains tenable. However, there are several unresolved anomalies regarding these and other technological distinctions (Hoffecker in press). It is admittedly difficult to explain why these two components nearly always occur in exactly the same Nenana valley sites, although in the case of Dry Creek the occupations are positioned directly at the point of the Healy terrace, which is the choice location along that bluff making it the most attractive setting for a band of hunter-gatherers. Despite this,

Thorson (2006) made an argument for post-depositional mixing at the Dry Creek site, but in my view presented no unequivocal evidence. Furthermore, arguments linking the

Nenana complex to the microblade bearing Chindadn assemblages in the Tanana valley

(Holmes in press; Holmes and Crass 2003) have yet to be fully accepted.

Future research should examine whether the Nenana complex is more frequently associated with the Bølling warm period, and the Denali complex with the Younger Dryas cold period. This could help explain the effects of long-term climate change on prehistoric technology. If seasonality is primarily responsible for these differences, additional research should focus on continued survey for new archeological sites from a variety of topographic settings which will provide key tests for the Annual Economic

Model presented in this dissertation. While the dichotomy between occupations that contain microblades and those that do not may greatly hinder interpretations based on the culture-historical model, when viewed as a function of seasonally organized technology in central and southcentral Alaska, the dichotomy is explained.

Conclusions

The first trace of hunter-gatherers in central Alaska dates to the post-glacial Bølling warm period (Holmes 2009). Although the deglaciation of southcentral Alaska was nearly 177

complete by the time the first human groups arrived in eastern Beringia, it took some time

before people crossed through the mountain passes of the Alaska Range and began using

the varied landscapes of southcentral Alaska. These events probably occurred during the

Younger Dryas when deteriorating conditions and dwindling megafauna nearly ended the

early human occupation of eastern Beringia and probably forced remaining groups out of

the interior and south of the Alaska Range (Chapter II).

The earliest known well-dated evidence for hunter-gatherers in southcentral Alaska was found at the alpine site Bull River II where biface production was undertaken in preparation for fall hunts in upland areas (Chapter IV). Further south in the Susitna River lowlands, two early Holocene components dated to the Preboreal and one middle Holocene occupation suggest the southerly expansion continued as hunter-gatherers became established in the area and probably began living year round south of the Alaska Range

(Chapter III).

In this dissertation, I assume that technological differences in central and southcentral Alaska are attributable to variability in human behavior rather than cultural differences. Lithic tools kits, site types, and site settings are generally consistent among

Denali complex occupations through time―hence, landuse and technological organization also changed little between the late Pleistocene and middle Holocene. This is why there are so many prehistoric sites in overlook settings on south facing slopes and why it is often difficult to judge if these are late Pleistocene or middle Holocene in age without precise dating techniques. Where there are differences between assemblages, they occur as a result of the activities undertaken, and some of these activities vary according to where they occur on the landscape. 178

Lithic analyses from four newly discovered sites in southcentral Alaska indicate that activities undertaken in different topographic zones were the result of seasonal discrepancies in raw material accessibility. Because patterns in site location, lithic reduction strategies, and toolkit maintenance differ in upland occupations from those in lowland settings, a seasonal model of technological organization and landuse was proposed. This is a rigorous model designed to move theory away from the cultural historic approach. It is also highly testable both with newly discovered and previously documented sites. Future research need not be guided by this specific model but should account for short and long-term changes in the environment as a driving force behind hunter-gatherer adaptations in the north. 179

APPENDIX

Table 7.1. Flaked debris frequencies at TCO and SRO. TCO-I TCO-II SRO Debitage Class n % n % n % Flake fragment 68 27.0 62 20.1 118 29.2 Complete flake 25 9.9 40 13.0 127 31.4 Bipolar flake 1 0.4 2 0.6 — — Table 7.2. Tool class frequencies at TCO and SRO. Blade-like-flake 2 0.8 2 0.6 1 0.2 TCO CI TCO CII SRO Microblade 10 4.0 13 4.2 — — Tool Class n % n % n % Prim. Cortical spall 6 2.4 11 3.6 11 2.7 Bifacial Point 1 6.7 — — — — Sec. Cortical spall 4 1.6 14 4.5 21 5.2 Biface unhafted 1 6.7 3 9.4 2 5.6 Retouch chip frag. 5 2.0 17 5.5 8 2.0 Flake tool 5 33.3 4 12.5 10 27.8 Retouch chip 42 16.7 66 21.4 30 7.4 Retouched flake — — 8 25 4 11.1 Bifacial thinning flake 52 20.6 56 18.2 62 15.3 Utilized Blade 1 6.7 — — 1 2.8 Burin spall 3 1.2 2 0.6 — — Burin 1 6.7 — — — — Angular shatter 25 9.9 13 4.2 25 6.2 End scraper 1 6.7 1 3.1 — — Cobble fragment — — 4 1.3 1 0.2 Side Scraper 2 13.3 4 12.5 2 5.6 Split cobble 2 0.8 2 0.6 — — Tci-thos 1 6.7 6 18.8 2 5.6 Worked chert cobble 1 0.4 1 0.3 — — Cobble Tool 1 6.7 6 18.8 6 16.7 Core fragment 2 0.8 — — — — Notch 1 6.7 — — — — Microblade core 2 0.8 — — — — — — — — Mulidirectional flake core — — 3 1.0 — — Chopper (biface) 5 13.9 Core rejuvenation flake 2 0.8 — — — — Chopper (uniface) — — — — 4 11.1 Total 252 100 308 100 404 100 Total 15 100 32 100 36 100

180

Table 7.3. Raw material type frequencies at TCO and SRO. TCO Component I TCO Component II Susitna River Overlook Debitage Tools Total Debitage Tools Total Debitage Tools Total Raw Material n % n % n % n % n % n % n % n % n % Basalt 75 29.8 5 33.3 80 30.0 125 40.6 10 31.3 135 39.7 295 73.0 21 58.3 316 71.8 Chalcedony 2 0.8 1 6.7 3 1.1 5 1.6 1 3.1 6 1.8 — — — — — — Chert 129 51.2 7 46.7 136 50.9 87 28.2 9 28.1 96 28.2 18 4.5 3 8.3 21 4.8 Obsidian 4 1.6 2 13.3 6 2.2 45 14.6 3 9.4 48 14.1 — — — — — — Sandstone 4 1.6 — — 4 1.5 5 1.6 2 6.3 7 2.1 65 16.1 7 19.4 72 16.4 Schist 28 11.1 — — 28 10.5 22 7.1 — — 22 6.5 — — — — — — Siltstone 6 2.4 — — 6 2.2 14 4.5 6 18.8 20 5.9 26 6.4 4 11.1 30 6.8 Slate 4 1.6 — — 4 1.5 5 1.6 1 3.1 6 1.8 — — — — — — Quartzite — — — — — — — — — — — — — — 1 2.8 1 0.2 Total 252 100 15 100 267 100 308 100 32 100 340 100 404 100 36 100 440 100

Table 7.4. Artifact size class frequencies at TCO and SRO. TCO Component I TCO Component II Susitna River Overlook Debitage Tools Total Debitage Tools Total Debitage Tools Total Size (cm) n % n % n % n % n % n % n % n % n % < 1 72 28.6 1 6.7 73 27.3 104 33.8 — — 104 30.6 44 10.9 — — 44 10.0 1-3 142 56.3 2 13.3 144 53.9 160 51.9 3 9.4 163 47.9 157 38.9 3 8.3 160 36.4 3-5 27 10.7 5 33.3 32 12.0 33 10.7 9 28.1 42 12.4 96 23.8 1 2.8 97 22.0 5-7 10 4.0 3 20.0 13 4.9 9 2.9 7 21.9 16 4.7 57 14.1 6 16.7 63 14.3 7-9 1 0.4 1 6.7 2 0.7 1 0.3 6 18.8 7 2.1 28 6.9 3 8.3 31 7.0 9-11 — — 2 13.3 2 0.7 — — 4 12.5 4 1.2 17 4.2 5 13.9 22 5.0 11-13 — — 1 6.7 1 0.4 — — 1 3.1 1 0.3 4 1.0 7 19.4 11 2.5 13-15 — — — — — — 1 0.3 2 6.3 3 0.9 1 0.2 2 5.6 3 0.7 15-17 — — — — — — — — — — — — — — 3 8.3 3 0.7 17-19 — — — — — — — — — — — — — — — — — — 19-21 — — — — — — — — — — — — — — 3 8.3 3 0.7 21-23 — — — — — — — — — — — — — — — — — — 23-26 — — — — — — — — — — — — — — 2 5.6 2 0.5 26-29 — — — — — — — — — — — — — — — — — — 29-31 — — — — — — — — — — — — — — 1 2.8 1 0.2 Total 252 100 15 100 267 100 308 100 32 100 340 100 404 100 36 100 440 100 181

Table 7.5. Raw material type frequencies at Bull River II and Costello Creek. Bull River II Costello Creek Debitage Tools Total Debitage Tools Total Raw Material n % n % n % n % n % n % Basalt 4 0.7 — — 4 0.7 9 2.1 1 25.0 10 2.3 Chalcedony 2 0.4 — — 2 0.3 — — — — — — Chert 16 2.8 — — 16 2.8 2 0.5 1 25.0 3 0.7 Sandstone 38 6.7 — — 38 6.6 22 5.2 — — 22 5.2 Siltstone 502 88.7 13 100 515 88.9 389 92.2 2 50.0 391 91.8 Claystone 4 0.7 — — 4 0.7 — — — — — — Total 566 100 13 100 579 100 422 100 4 100 426 100

Table 7.6. Artifact size class frequencies at Bull River II and Costello Creek. Bull River II Costello Creek Debitage Tools Total Debitage Tools Total Size (cm) n % n % n % n % n % n % < 1 17 3.0 — — 17 2.9 7 1.7 — — 7 1.6 1-3 323 57.1 — — 323 55.8 271 64.2 2 50.0 273 64.1 3-5 160 28.3 — — 160 27.6 106 25.1 1 25.0 107 25.1 5-7 52 9.2 12 92.3 64 11.1 26 6.2 1 25.0 27 6.3 7-9 9 1.6 1 7.7 10 1.7 9 2.1 — — 9 2.1 9-11 5 0.9 — — 5 0.9 2 0.5 — — 2 0.5 11-13 — — — — — — 1 0.2 — — 1 0.2 Total 566 100 13 100 579 100 422 100 4 100 426 100

182

Table 7.7. Flaked debris frequencies at Bull River II and Costello Creek. Bull River Costello Total II Creek Table 7.8. Tool class frequencies at Bull River II and Costello Creek. Debitage Class n % n % n % Bull River II Costello Creek Total

Flake fragment 201 35.5 140 33.2 341 34.5 Tool Type n % n % n % Complete flake 165 29.2 141 33.4 306 31.0 Biface unhafted 5 38.5 — — 5 29.4 Blade-like-flake 2 0.4 2 0.5 4 0.4 Prim. Cortical spall 94 16.6 46 10.9 140 14.2 Flake tool 3 23.1 1 25.0 4 23.5 Sec. Cortical spall 25 4.4 13 3.1 38 3.8 Retouched flake 3 23.1 1 25.0 4 23.5 Retouch chip frag. 2 0.4 1 0.2 3 0.3 Side Scraper — — 1 25.0 1 5.9 Retouch chip 6 1.1 3 0.7 9 0.9 Tci-thos 1 7.7 — — 1 5.9 Bifacial thinning flake 42 7.4 56 13.3 98 9.9 Notch 1 7.7 1 25.0 2 11.8 Angular shatter 24 4.2 16 3.8 40 4.0 Total 13 100 4 100 17 100 Worked chert cobble — — 1 0.2 1 0.1 Core fragment 5 0.9 3 0.7 8 0.8 Total 566 100 422 100 988 100

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (methods of rejection). The following table contains the results of the radiocarbon data evaluation criteria from Chapter V. Dates rejected by the Method I criteria are noted by the superscript a. Dates rejected by Method II are noted by the superscript b. Dates considered aberrant by the original investigators are indicated by an asterisk and were rejected in both methods. In cases where components were represented by multiple radiocarbon samples within a 100 year interval (e.g. 9800-9900), the date with the smallest standard deviation was selected and remaining dates were rejected. Rejection based on this criterion is noted in by the superscript c. Only dates lacking a superscript notation and those labeled b were used in the less conservative Method I dataset. Method II differed in that it had a more restricted σ tolerance and includes only those dates which completely lack a superscript notation. 183

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades. Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Annie Lake, Yukon Yes Beta-57946 6230+70 7130+100 charcoal Hare 1995; Mason et al. 2001 Annie Lake, Yukon c Yes Beta-57947 6260+80 7160+110 charcoal Hare 1995; Mason et al. 2001 Annie Lake, Yukon Yes Beta-57944 7160+70 7990+70 charcoal Hare 1995; Mason et al. 2001 Bedwell No CAMS-11032 10490+70 12450+170 associated with lanceolate point Hamilton and Goebel 1999:191; Reanier 1995 Broken Mammoth I A No WSU-4267 2040+65 2020+90 charcoal Hamilton and Goebel 1999; Holmes et al. 1996:315 Broken Mammoth I A b No UGA-6255D 2815+180 2990+210 charcoal Hamilton and Goebel 1999; Holmes et al. 1996:315 Broken Mammoth I B c Yes WSU-4458 4525+90 5170+140 charcoal, W scatter, probable hearth Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth I B Yes WSU-4456 4540+90 5190+140 charcoal, E. concentration Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth I B b, c Yes WSU-4457 4545+90 5200+150 charcoal, W scatter, probable hearth Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth I B Yes WSU-4350 4690+110 5410+140 central hearth Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth II b Yes UGA-6281D 7201+205 8040+210 charcoal, N. hearth, split WSU-4264 Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth II Yes WSU-4264 7600+140 8400+150 charcoal, N. hearth, split UGA-6281D Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth II Yes WSU-4208 7700+80 8500+70 charcoal, W. hearth Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth II Yes AA 19323 8015+75 8870+120 charcoal, hearth feature Holmes 2001:157 Broken Mammoth III a Yes WSU-4266 9310+65 10500+100 combined, upper stringer, middle p. Krasinski and Yesner 2008 Broken Mammoth III a Yes UGA-6256D 9690+960 11140+1260 split with WSU-4263 Krasinski and Yesner 2008 Broken Mammoth III c Yes WSU-4263 10270+110 12050+250 central hearth, middle/middle paleosol Krasinski and Yesner 2008 Broken Mammoth III Yes CAMS-5357 10290+70 12100+180 charcoal, E. hearth, middle/middle p. Krasinski and Yesner 2008 Broken Mammoth III b Yes WSU-4019 10790+230 12670+280 charcoal scatter Krasinski and Yesner 2008 Broken Mammoth IV A No CAMS-7203 11040+80 12990+70 large mammal bone; btw paleosol Hamilton and Goebel 1999:187; Holmes et al. 1996:316 Broken Mammoth IV A a No UGA-6257D 11040+260 12990+210 charcoal Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth IV A c No CAMS-7204 11060+90 13000+80 wapiti bone; btw paleosols Hamilton and Goebel 1999:187; Holmes et al. 1996:317 Broken Mammoth IV B b No WSU-4265 11280+190 13180+170 charcoal, middle/lower paleo, combined Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth IV B No CAMS-5358 11420+70 13290+70 charcoal, E. hearth, middle/lower p. Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth IV B No CAMS-8261 11500+80 13360+80 swan bone, middle/lower paleosol Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth IV B c No WSU-4262 11510+120 13400+130 charcoal, central hearth, middle/lower Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth IV B c No NA 11540+148 13430+150 ivory collagen Holmes 2001:158 Broken Mammoth IV C b No WSU-4351 11770+210 13630+210 charcoal Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth IV C b, c No WSU-4364 11770+220 13640+220 charcoal Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Broken Mammoth IV C* No CAMS-9898 15830+70 19050+100 mammoth ivory Hamilton and Goebel 1999; Holmes et al. 1996; Yesner et al. 1992 Butte Lake II b Yes Beta 10751 5030+200 5800+220 charcoal feature 11 Betts 1987:42 Butte Lake II a Yes BETA 10750 6390+580 7200+590 charcoal, feature 1 Betts 1987:42 Bull River II c No Beta-234548 10310+50 12160+130 charcoal Wygal 2007c Bull River II No Beta-234549 10350+50 12220+120 charcoal Wygal 2007c Bull River II No Beta-234746 10460+50 12420+160 charcoal Wygal 2007c Bull River II c No Beta-234747 10490+50 12480+130 charcoal Wygal 2007c 184

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (continued). Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Campus Yes Beta-7025 2725+125 2870+140 charcoal Mobley 1991, 1996:301 Campus b Yes Beta-4260 2860+180 3030+210 charcoal Mobley 1991, 1996:301 Campus Yes Beta-6829 3500+140 3790+180 charcoal Mobley 1991, 1996:301 Campus Yes Beta-97212 6850+70 7700+70 charcoal Mason et al. 2001; Mobley 1991 Carlo Creek I * No WSU-1727 5120+265 5890+290 soil humic acid, lower paleosol Bowers 1980; Hamilton and Goebel 1999:190 Carlo Creek I b No WSU-1700 8400+200 9350+230 charcoal, hearth I Bowers 1980; Hamilton and Goebel 1999:190 Carlo Creek I a No GX-5132 8690+330 9760+400 charcoal, hearth II Bowers 1980; Hamilton and Goebel 1999:191 Carlo Creek I a No AA-75050 9647+60 11000+150 charcoal, hearth 1 (split) Bowers and Reuther 2008 Carlo Creek I a No AA-75051 9763+50 11190+40 humates, hearth 1 (split) Bowers and Reuther 2008 Carlo Creek I c No AA-75049 9872+65 11320+90 bone Bowers and Reuther 2008 Carlo Creek I No AA-75051 9902+50 11340+80 charcoal, hearth 1 (split) Bowers and Reuther 2008 Carlo Creek I No AA-75052 9979+50 11460+130 charcoal, hearth 2 Bowers and Reuther 2008 Carlo Creek I a No AA-75053 10035+50 11550+140 humates, hearth 2 Bowers and Reuther 2008 Carlo Creek I *,a No GX-5131 10040+435 11660+680 charcoal, hearth I Bowers 1980; Hamilton and Goebel 1999:190 Chugwater II No Beta-18509 8960+130 10020+190 charcoal Lively 1996:309; Mason et al. 2001 Chugwater II b No Beta-19498 9460+160 10780+260 charcoal Lively 1996:309; Mason et al. 2001 Delta R. Overlook V No GX-6750 2285+145 2340+210 hearth stain Holmes 2001:159; Holmes and Bacon 1982 Delta R. Overlook III b Yes GX-6752 3980+150 4460+230 paleosol 4, 35 cm below bison bone Holmes 2001:159; Holmes and Bacon 1982 Delta R. Overlook II b Yes GX-6749 6675+175 7560+150 just above CZ 2 Holmes and Bacon 1982; Mason et al. 2001 Delta R. Overlook II b Yes GX-6751 7190+200 8020+200 just below CZ 2 Holmes and Bacon 1982; Mason et al. 2001 Delta R. Overlook I a Yes NA 8555+380 9590+490 date from just above CZ I Bacon and Holmes 1980; Holmes 2001:159; Potter 2005:Table 12 Dry Creek IV No SI-2332 3430+75 3700+100 charcoal Hoffecker et al. 1996a; Powers and Hoffecker 1989:270 Dry Creek IV No SI-1934 3655+60 3990+90 charcoal Hoffecker et al. 1996a; Powers and Hoffecker 1989:270 Dry Creek IV No SI-1937 4670+95 5400+130 charcoal Hoffecker et al. 1996a; Powers and Hoffecker 1989:270 Dry Creek II Yes AA-11730 8915+70 10030+120 charcoal, paleosol 1 Hoffecker et al. 1996a; Powers and Hoffecker 1989:270 Dry Creek II Yes AA-11733 9340+95 10540+140 charcoal, paleosol 2 Hoffecker et al. 1996a Dry Creek II b, c Yes SI-2329 9340+195 10640+300 charcoal, paleosol 2 Hoffecker et al. 1996a Dry Creek II Yes AA-11732 9690+75 11030+150 charcoal, paleosol 2 Hoffecker et al. 1996a Dry Creek II Yes AA-11727 10060+75 11610+190 charcoal, paleosol 1 Hoffecker et al. 1996a; Powers and Hoffecker 1989:270 Dry Creek II Yes AA-11731 10540+70 12550+140 charcoal, paleosol 2 Hoffecker et al. 1996a Dry Creek II Yes AA-11728 10615+100 12600+170 charcoal, paleosol 1 Hoffecker et al. 1996a Dry Creek II a, c Yes SI-1561 10690+250 12520+330 charcoal, paleosol 1 Hoffecker et al. 1996a; Powers and Hoffecker 1989:270 Dry Creek I No SI-2880 11120+85 13040+80 charcoal Hoffecker et al. 1996a; Powers and Hoffecker 1989:270 Dry Creek * ? SI-1935A 10600+500 12240+660 charcoal, paleosol 3 Hamilton and Goebel 1999:186; Thorson and Hamilton 1977 Dry Creek * ? SI-1936 12080+1025 14310+1280 charcoal, paleosol 2 Hamilton and Goebel 1999:186; Thorson and Hamilton 1977 Dry Creek * ? SI-1544 19050+1500 22800+1850 charcoal, paleosol 3 Hamilton and Goebel 1999:186; Thorson and Hamilton 1977 185

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (continued). Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Dry Creek * ? SI-1938 23930+9300 26950 charcoal, paleosol 2 Hamilton and Goebel 1999:186; Thorson and Hamilton 1977 Gallagher Flint Station Yes Beta-97211 6960+90 7800+90 charcoal from microblade component Ferguson 1997; Hamilton and Goebel 1999:191; Mason et al. 2001 Gallagher Flint Station b Yes SI-974 10540+150 12460+250 charcoal from microblade component Dixon 1975; Ferguson 1997; Hamilton and Goebel 1999:191 Gerstle River V No WSU-4888 7600+140 8400+150 charred material Potter 2005:Table5.1, 571 Gerstle River V No Beta-98433 8380+50 9400+70 charred material Potter 2005:Table5.1, 571 Gerstle River IV Yes Beta-167396 8660+40 9620+50 hearth feature 7 Potter 2005:Table5.1, 571 Gerstle River III Yes Beta-191558 8760+40 9770+90 hearth feature 14 Potter 2005:Table5.1, 571 Gerstle River III c Yes Beta-183109 8820+50 9920+150 hearth feature 16 Potter 2005:Table5.1, 571 Gerstle River III c Yes Beta-181678 8830+50 9940+150 hearth feature 12 Potter 2005:Table5.1, 571 Gerstle River III c Yes Beta-133750 8860+70 9960+160 hearth feature 1 Potter 2005:Table5.1, 571 Gerstle River III Yes Beta-167397 8890+40 10040+100 hearth feature 5 Potter 2005:Table5.1, 571 Gerstle River III c Yes Beta-181679 8900+40 10040+100 hearth feature 13 Potter 2005:Table5.1, 571 Gerstle River III Yes Beta-167399 8910+40 10050+100 hearth feature 10 Potter 2005:Table5.1, 571 Gerstle River III c Yes Beta-167395 8950+40 10080+110 hearth feature 3 Potter 2005:Table5.1, 571 Gerstle River III c Yes AA-51254 9030+70 10140+120 hearth feature 9 Potter 2005:Table5.1, 571 Gerstle River III Yes Beta-183108 9080+50 10250+40 hearth feature 18 Potter 2005:Table5.1, 571 Gerstle River III Yes Beta-167398 9130+40 10310+60 hearh feature 8 Potter 2005:Table5.1, 571 Gerstle River III c Yes AA-51253 9130+70 10330+90 hearth feature 17 Potter 2005:Table5.1, 571 Gerstle River II Yes Beta-183110 9400+50 10630+60 hearth feature 17 Potter 2005:Table5.1, 571 Gerstle River II Yes Beta-134098 9510+50 10870+150 hearth feature 2 Potter 2005:Table5.1, 571 Gerstle River I No Beta-133751 9740+50 11170+50 charred material, paleosol Potter 2005:Table5.1, 571 Healy Lake Village L 1 a Yes GX-1945 modern charcoal Cook 1996:326-327 Healy Lake Village L 1 a Yes GX-2166 455+130 470+130 charcoal Cook 1996:326-327 Healy Lake Village L 1 a, c Yes GAK-1886 900+90 840+80 charcoal Cook 1996:326-327 Healy Lake Village L 2 a Yes Beta-76058 380+50 430+70 plant remains Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 2 a Yes GX-2160 905+90 840+80 charcoal Cook 1996:326-327 Healy Lake Village L 2 a Yes GAK-1887 1360+80 1280+80 charcoal Cook 1996:326-327 Healy Lake Village L 2 a Yes GX-2168 1655+180 1600+190 charcoal Cook 1996:326-327 Healy Lake Village L 2 a Yes GX-2169 2875+140 3050+180 charcoal Cook 1996:326-327 Healy Lake Village L 2 a, c Yes GX-2169 2880+140 3050+180 charcoal Cook 1996:326-327 Healy Lake Village L 2 a Yes GX-2165 3580+140 3900+190 charcoal Cook 1996:326-327 Healy Lake Village L 2 a Yes AU-4 3655+426 4050+550 charcoal Cook 1996:326-327 Healy Lake Village L 2 a Yes Beta-76068 4460+60 5110+130 plant remains Cook 1996:326-327 Healy Lake Village L 3 a Yes Beta-76059 1790+50 1730+80 plant remains Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 3 a Yes GX-2176 2660+100 2740+140 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 3 a, c Yes DIC-3065 2680+150 2770+210 charcoal Cook 1996:326-327 186

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (continued). Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Healy Lake Village L 3 a Yes Beta-76061 3350+50 3590+70 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 4 a Yes GX-2161 2150+180 2150+210 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 4 a Yes Beta-76063 3020+50 3230+90 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 4 a Yes GX-2163 4010+110 4510+190 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 4 a, c Yes GX-1340 8960+150 10020+210 bone Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 5 a Yes Beta-76069 5000+60 5760+90 plant remains Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 6 a Yes Beta-76064 5110+90 5850+100 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 6 a Yes Beta-76062 7920+90 8790+140 soil Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 6 a Yes Beta-76071 10410+60 12350+180 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 6 a Yes Beta-76067 11100+60 13020+70 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 6 a Yes Beta-76060 11410+60 13280+60 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a, c Yes GX-2171 8655+280 9740+360 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a Yes GX-2170 8680+240 9780+300 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a Yes Beta-76070 8990+60 10100+120 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a Yes AU-1 9245+213 10500+280 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a Yes GX-2174 9895+210 11440+360 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a Yes SI-737 10150+210 11840+400 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a Yes Beta-76066 10290+60 12110+150 charcoal Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 7 a Yes Beta-76065 11550+50 13400+70 plant remains Cook 1996:326-327; Hamilton and Goebel 1999:188 Healy Lake Village L 8 a Yes GX-1341 11090+170 13040+130 bone Cook 1996:326-327 Healy Lake Village L 9 a No GX-2159 6645+280 7510+270 charcoal Cook 1996:326-327 Healy Lake Village L 9 a No SI-738 8210+155 9160+210 charcoal Cook 1996:326-327 Healy Lake Village L 9 a No AU-2 9401+528 10780+780 charcoal Cook 1996:326-327 Healy Lake Village L 10 a No GX-2175 8465+360 9480+470 charcoal Cook 1996:326-327 Healy Lake Village L 10 a No SI-739 10040+210 11680+350 charcoal Cook 1996:326-327 Healy Lake Village L 10 a No AU-3 10434+279 12200+440 charcoal Cook 1996:326-327 Healy Lake Village L 10 a No GX-1944 10500+280 12270+420 charcoal Cook 1996:326-327 Hilltop a ? GaK-4924 6160+130 7050+160 soil organics Hamilton and Goebel 1999:191; Reanier 1995 Hilltop No CAMS-11034 10360+60 12240+130 charcoal Hamilton and Goebel 1999:191; Reanier 1995 Little Panguingue Creek Yes AA-1699 1825+68 1760+90 charcoal, hearth feature Hoffecker and Powers 1996:371; Powers and Hoffecker 1989:270 Mead CZ I No WSU-4348 1430+60 1360+50 charcoal Hamilton and Goebel 1999:188 Mead CZ II No WSU-4349 4050+140 4550+210 charcoal, middle paleosol Hamilton and Goebel 1999:188 Mead CZ II b No Beta-59115 6070+170 6950+210 charcoal Hamilton and Goebel 1999:188 Mead CZ II * No Beta-59117 9220+370 10460+530 charcoal, forest soil Hamilton and Goebel 1999:188 Mead CZ III * No WSU-4261 7620+100 8440+90 charcoal, bulk sample Hamilton and Goebel 1999:188 187

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (continued). Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Mead CZ III No CAMS-5197 10410+80 12350+190 charcoal, middle paleosol Hamilton and Goebel 1999:188 Mead CZ III c No CAMS-4876 10460+110 12400+220 charcoal, middle paleosol Hamilton and Goebel 1999:188 Mead CZ III-IV b No WSU-4425 10760+170 12670+210 charcoal, lower-middle paleosol Hamilton and Goebel 1999:188 Mead CZ IV No CAMS-5198 11560+80 13430+100 charcoal, lower paleosol Hamilton and Goebel 1999:188 Mead CZ IV No CAMS-4877 11600+80 13470+110 charcoal, lower paleosol Hamilton and Goebel 1999:188 Mead CZ IV* No CAMS-17408 17370+90 20510+140 collagen Hamilton and Goebel 1999:188 Mesa, East Ridge b No Beta-120397 8820+230 9900+280 hearth Kunz et al. 2003:20-21 Mesa, East Ridge No Beta-125996 9330+40 10550+60 hearth Kunz et al. 2003:20-21 Mesa, East Ridge c No Beta-120400 9740+50 11170+50 hearth Kunz et al. 2003:20-21 Mesa, East Ridge No Beta-130577 9780+40 11210+30 hearth Kunz et al. 2003:20-21 Mesa, East Ridge No GX-26567-AMS 9930+40 11370+90 hearth Kunz et al. 2003:20-21 Mesa, East Ridge No Beta-125998 10030+40 11540+130 hearth Kunz et al. 2003:20-21 Mesa, East Ridge c No Beta-125997 10080+40 11630+140 hearth Kunz et al. 2003:20-21 Mesa, East Ridge a No GX-26461 12240+610 14400+770 hearth Kunz et al. 2003:20-21 Mesa, Locality A No Beta-125995 9160+140 10370+160 hearth Kunz et al. 2003:20-21 Mesa, Locality A b No Beta-140199 9500+190 10810+270 hearth Kunz et al. 2003:20-21 Mesa, Locality A No Beta-96065 9810+110 11250+180 hearth Kunz et al. 2003:20-21 Mesa, Locality A b, c No Beta-96067 9850+150 11350+260 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-119100 10000+50 11490+130 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-118584 10040+50 11560+140 hearth Kunz et al. 2003:20-21 Mesa, Locality A No Beta-118583 10050+50 11580+150 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-84650 10080+50 11630+160 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-96068 10080+120 11670+250 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-95913 10080+60 11640+170 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-96066 10090+110 11680+240 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-118582 10100+50 11690+170 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-142262 10120+50 11750+160 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-95914 10130+60 11760+170 hearth Kunz et al. 2003:20-21 Mesa, Locality A No Beta-118585 10130+50 11780+150 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-96069 10150+130 11770+270 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-118581 10170+50 11860+120 hearth Kunz et al. 2003:20-21 Mesa, Locality A No Beta-95600 10230+60 11950+130 hearth Kunz et al. 2003:20-21 Mesa, Locality A c No Beta-96070 10260+110 12030+250 hearth Kunz et al. 2003:20-21 Mesa, Locality B * No DIC-1589 7620+95 8440+90 combined hearths (n=3); Locality B Kunz et al. 2003:20-21; Hamilton and Goebel 1999:173 Mesa, Locality B c No Beta-57429 9900+70 11390+130 hearth Kunz et al. 2003:20-21 188

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (continued). Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Mesa, Locality B c No Beta-69899 9900+80 11410+150 hearth Kunz et al. 2003:20-21 Mesa, Locality B c No Beta-55284 9930+80 11440+150 hearth Kunz et al. 2003:20-21 Mesa, Locality B No Beta-84649 9980+60 11470+140 hearth Kunz et al. 2003:20-21 Mesa, Locality B c No Beta-55285 10000+80 11520+170 hearth Kunz et al. 2003:20-21 Mesa, Locality B c No Beta-69900 10050+90 11600+200 hearth Kunz et al. 2003:20-21 Mesa, Locality B No Beta-52606 10060+70 11610+180 hearth Kunz et al. 2003:20-21 Mesa, Locality B c No Beta-50428 10090+85 11670+220 hearth Kunz et al. 2003:20-21 Mesa, Locality B No Beta-55283 10240+80 11980+170 hearth Kunz et al. 2003:20-21 Mesa, Locality B * No Beta-57430 11190+70 13090+80 hearth, old wood? Kunz et al. 2003:20-21; Hamilton and Goebel 1999:176 Mesa, Locality B * No Beta-55286 11660+80 13530+110 hearth, old wood? Kunz et al. 2003:20-21; Hamilton and Goebel 1999:176 Mesa, Saddle a No Beta-140198 9480+710 10920+1010 hearth Kunz et al. 2003:20-21 Mesa, Saddle No Beta-36805 9730+80 11060+160 hearth Kunz et al. 2003:20-21 Mesa, Saddle c No Beta-120793 9800+60 11220+50 hearth Kunz et al. 2003:20-21 Mesa, Saddle No Beta-120399 9860+50 11280+50 hearth Kunz et al. 2003:20-21 Mesa, Saddle No Beta-120398 9920+50 11380+100 hearth Kunz et al. 2003:20-21 Mesa, Saddle c No Beta-50430 9945+75 11450+150 hearth Kunz et al. 2003:20-21 Mesa, Saddle c No Beta-133354 9950+60 11440+140 hearth Kunz et al. 2003:20-21 Mesa, Saddle c No Beta-55282 9990+80 11510+170 hearth Kunz et al. 2003:20-21 Mesa, Saddle c No Beta-69898 10070+60 11620+170 hearth Kunz et al. 2003:20-21 Mesa, Saddle No Beta-142261 10080+50 11630+160 hearth Kunz et al. 2003:20-21 Mesa, Saddle No Beta-133353 10180+60 11860+130 hearth Kunz et al. 2003:20-21 Mesa, Saddle a No Beta-50429 10980+280 12890+280 hearth Kunz et al. 2003:20-21 Moose Creek CZ I No Beta-96627 11190+60 13090+80 hearth Powers and Hoffecker 1989:270; Hoffecker 1996 Onion Portage, Akmak * Yes K-1583 9570+150 10900+210 redeposited bone Anderson 1988; Hamilton and Goebel 1999:191 Onion Portage, Band 8 * Yes P-1111 7180+90 8020+100 charcoal, Kobuk Complex Anderson 1988; Hamilton and Goebel 1999:191 Onion Portage, Band 8 * Yes P-1111A 7320+100 8160+110 charcoal, Kobuk Complex Anderson 1988; Hamilton and Goebel 1999:191 Onion Portage, Band 8 c Yes P-1076 7900+100 8770+150 charcoal, Kobuk Complex Anderson 1988; Hamilton and Goebel 1999:191 Onion Portage, Band 8 Yes P-984A 7920+100 8790+150 charcoal, Kobuk Complex Anderson 1988; Hamilton and Goebel 1999:191 Onion Portage, Band 8 Yes P-985 8100+100 9010+190 charcoal, Kobuk Complex Anderson 1988; Hamilton and Goebel 1999:191 Onion Portage, Band 8 a, c Yes P-985 8195+280 9090+340 charcoal, Kobuk Complex Anderson 1988; Hamilton and Goebel 1999:191 Otter Falls b Yes GSC-942 4590+150 5250+210 ? Workman 1978:186 Owl Ridge I * No Beta-11079 2380+90 2500+150 charcoal? Hoffecker et al. 1996b; Phippen 1988 Owl Ridge I * No Beta-5416 9060+410 10260+550 charcoal Hoffecker et al. 1996b; Phippen 1988 Owl Ridge I b No Beta-11209 11340+150 13220+140 charcoal Hoffecker et al. 1996b; Phippen 1988 Owl Ridge II * No Beta-11081 2470+120 2550+150 charcoal? Hoffecker et al. 1996b; Phippen 1988 189

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (continued). Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Owl Ridge II a No Beta-11436 7660+100 8470+90 soil organics Hoffecker et al. 1996b; Phippen 1988 Owl Ridge II a No Beta-5418 8130+140 9050+230 soil organics Hoffecker et al. 1996b; Phippen 1988 Owl Ridge II a No GX-6283 9325+305 10610+420 soil organics Hoffecker et al. 1996b; Phippen 1988 Owl Ridge III * No Beta-11082 1480+180 1420+190 charcoal Hoffecker et al. 1996b; Phippen 1988 Owl Ridge III a No D-3070 6900+265 7770+230 charcoal Hoffecker et al. 1996b; Phippen 1988 Owl Ridge III a No GX-13009 7035+389 7900+360 charcoal Hoffecker et al. 1996b; Phippen 1988 Owl Ridge a ? Beta-11080 440+70 460+80 charcoal Hoffecker et al. 1996b; Phippen 1988 Owl Ridge a ? D-3071 930+50 860+60 charcoal, upper sand Hoffecker et al. 1996b; Phippen 1988 Panguingue Creek I * No AA-1687 8170+120 9130+170 carbonized sediment? Goebel and Bigelow 1996:369; Powers and Hoffecker 1989:270 Panguingue Creek I No GX-17457 9836+62 11270+60 charcoal Goebel and Bigelow 1996:369 Panguingue Creek I No AA-1686 10180+130 11830+280 charcoal Goebel and Bigelow 1996:369; Powers and Hoffecker 1989:270 Panguingue Creek I a No A-6744 13535+400 16080+550 basal soil organics Hamilton and Goebel 1999:187 Panguingue Creek II b Yes Beta-15094 7130+180 7960+180 charcoal Goebel and Bigelow 1996:369; Hamilton and Goebel 1999:187 Panguingue Creek II a Yes AA-1688 7430+270 8270+270 charcoal Goebel and Bigelow 1996:369; Powers and Hoffecker 1989:270 Panguingue Creek II a Yes GX-13012 7595+405 8510+450 charcoal Goebel and Bigelow 1996:369; Powers and Hoffecker 1989:270 Panguingue Creek II b Yes AA-1689 8600+200 9690+270 charcoal Goebel and Bigelow 1996:369; Powers and Hoffecker 1989:270 Panguingue Creek III No GX-13011 4510+95 5150+150 charcoal Goebel and Bigelow 1996:369; Powers and Hoffecker 1989:270 Putu a No WSU-1318 8454+130 9410+130 soil organics Alexander 1987; Hamilton and Goebel 1999:191 Putu No Beta-69901 8810+60 9900+170 hearth feature 9 Alexander 1987; Hamilton and Goebel 1999:191 Putu * No SI-2382 11470+500 13450+510 hearth Alexander 1987; Hamilton and Goebel 1999:191 Susitna River Overlook No Beta-208284 8160+50 9130+80 paleosol charcoal Wygal and Goebel 2006 Swan Point I B Yes WSU-4523 1220+70 1160+90 Betula sp. charcoal Holmes et al. 1996: 321 Swan Point I B Yes WSU-4524 1570+70 1480+70 Picea sp. charcoal Holmes et al. 1996: 321 Swan Point I B Yes WSU-4522 1670+60 1590+90 charcoal Holmes et al. 1996: 321 Swan Point I B Yes WSU-4521 1750+80 1690+100 charcoal/resin Holmes et al. 1996: 321 Swan Point II Yes WSU-4426 7400+80 8220+100 charcoal Holmes et al. 1996: 321 Swan Point III Yes Beta-56666 10230+80 11950+170 charcoal Holmes et al. 1996: 321 Swan Point IV A Yes Beta-71372 11660+60 13530+90 Salix sp. Holmes and Crass 2003; Holmes et al. 1996: 321 Swan Point IV A c Yes Beta-56667 11660+70 13530+100 Salix sp. hearth Holmes and Crass 2003; Holmes et al. 1996: 321 Swan Point IV A Yes AA-19322 11770+140 13620+150 substance from a MB core/hearth lens Morland 2007; Holmes and Crass 2003 Swan Point IV B Yes NA 12040+40 13900+70 charcoal Holmes and Crass 2003 Swan Point IV B c Yes NSRL-2001 12060+70 13920+90 mammoth ivory collagen Holmes et al. 1996: 321 Swan Point IV B Yes NA 12110+50 13970+80 charcoal Holmes and Crass 2003 Swan Point IV B Yes NA 12360+60 14380+190 charcoal Holmes and Crass 2003 Teklanika West I Yes GX-18518 7130+98 7960+100 charcoal Goebel 1996:341; Hamilton and Goebel 1999:190 190

Table 7.9. Evaluation of dated components in eastern Beringia and the presence or absence of microblades (continued). Site\Component M.B. Lab # 14C B.P. cal B.P. (1σ) Note Reference Teklanika West II Yes Beta-5340 3310+100 3560+110 charcoal Goebel 1996:341; Hamilton and Goebel 1999:190 Teklanika West II Yes GX-18517 5340+90 6120+110 charcoal Goebel 1996:341; Hamilton and Goebel 1999:190 Tingmiakpuk Yes Beta-49166 7060+70 7880+70 caribou bone Mason et al. 2001 Tingmiakpuk Yes Beta-49163 8435+85 9420+90 caribou bone Mason et al. 2001 Trail Creek Caves b Yes K-980 9070+150 10200+230 caribou bone Mason et al. 2001 Trail Creek Caves a Yes K-1327 11360+280 13290+260 mammoth scapula cave B Hamilton and Goebel 1999:191; Larsen 1968 Teklanika West II Yes Beta-5340 3310+100 3560+110 charcoal Goebel 1996:341; Hamilton and Goebel 1999:190 Trail Creek Caves a Yes K-1327 13070+280 15470+390 bison cal. cave 9 Hamilton and Goebel 1999:191; Larsen 1968 Trail Creek Caves a Yes Beta-20027 14270+950 17040+1260 mammoth vert. cave B Hamilton and Goebel 1999:191; Larsen 1968 Trail Creek Caves a Yes K-1210 15750+350 19030+300 horse scapula cave 9 Hamilton and Goebel 1999:191; Larsen 1968 Trapper Cr. Overlook, 4A Yes BETA208282 5140+40 5870+80 concentrated charcoal, paleosol 2 Wygal and Goebel 2006 Trapper Cr. Overlook, 4A c Yes AA72199 5186+42 5950+40 concentrated charcoal, paleosol 2 Wygal and Goebel 2006 Trapper Cr. Overlook, 4B Yes Beta-208283 7930+40 8800+120 concentrated charcoal, paleosol 2 Wygal and Goebel 2006 Tuluaq Hill * No NA 7950+40 8830+110 charcoal Rasic and Gal 2000:76 Tuluaq Hill c No Beta-122323 11110+80 13030+80 shallow feature Rasic 2003 Tuluaq Hill c No Beta-159913 11120+40 13030+60 charcoal Rasic and Gal 2000:76 Tuluaq Hill No Beta-159915 11160+40 13060+60 shallow feature Rasic 2003 Tuluaq Hill c No Beta-122322 11180+80 13080+90 charcoal Rasic and Gal 2000:76 Tuluaq Hill c No Beta-133393 11200+40 13110+60 charcoal Rasic and Gal 2000:76 Tuluaq Hill c No Beta-159914 11200+40 13110+60 shallow feature Rasic 2003 Usibelli a Yes GX-13013 3195+295 3420+370 charcoal Powers and Hoffecker 1989:270 Walker Road I b No AA-1683 11010+230 12970+180 charcoal Hamilton and Goebel 1999:186; Powers and Hoffecker 1989:270 Walker Road I b No AA1681 11170+180 13090+140 charcoal Hamilton and Goebel 1999:186; Powers and Hoffecker 1989:270 Walker Road I No AA-2264 11300+120 13180+110 charcoal Hamilton and Goebel 1999:186; Powers and Hoffecker 1989:270 Note dates rejected by primary investigators (*), Method I (a), Method II (b), overlapping dates from single components within a 100 year interval (c), italics indicate AMS dates, and M.B. refers to microblades. 191

REFERENCES CITED

Abbott, M.B, B.P. Finney, M.E. Edwards, and K.R. Kelts 2000 Lake-Level Reconstruction and Paleohydrology of Birch Lake, Central Alaska, Based on Seismic Reflection Profiles and Core Transects. Quaternary Research 53(2):154-166.

Ackerman, R.E. 1992 Earliest Stone Industries of the North Pacific Coast of North America. Arctic Anthropology 29(2):18-27.

1996a Ground Hog Bay, Site 2. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 424-430. University of Chicago Press, Chicago.

1996b Early Maritime Culture Complexes of the Northern Northwest Coast. In Early Human Occupation in British Columbia, edited by R.B. Carlson, pp. 123-132. University of British Columbia Press, Vancouver.

1996c Spein Mountain. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 456-460. University of Chicago Press, Chicago.

1996d Lime Hills Cave 1. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 470-477. University of Chicago Press, Chicago.

in press Microblade Assemblages in Southwestern Alaska: An Early Holocene Adaptation. In From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia, edited by T. Goebel and I. Buvit. Texas A&M University Press, College Station.

Ackerman, R.E., T.D. Hamilton and R. Stuckenrath 1979 Early Cultural Complexes on the Northern Northwest Coast. Canadian Journal of Archaeology 3:195-210.

Ackerman, R.E., K.C. Reid, J.D. Gallison and M.E. Roe 1985 Archaeology of Heceta Island: A Survey of 16 Timber Harvest Units in the Tongass National Forest, Southeastern Alaska. Washington State University. Report No. 53-0109-5-00181, U.S. Department of Agriculture, Forest Service, Ketchikan, Alaska.

Ager, T.A. 1975 Late Quaternary Environmental History of the Tanana Valley, Alaska. Institute for Polar Studies Report 54, Ohio State University. 192

1983 Holocene Vegetation History of Alaska. In Late Quaternary Environments of the United States, The Holocene Vol 2, edited by H. Wright Jr., pp. 128-141. University of Minnesota Press, Minneapolis.

1989 History of Late Pleistocene and Holocene Vegetation in the Copper River Basin, Southcentral Alaska. In Late Cenozoic History of the Interior Basins of Alaska and the Yukon, edited by L. Carter, T. Hamilton, J. Galloway, pp. 89-92. US Geological Survey Circular 1026.

Aitken, M. J. and S.G.E. Bowman 1975 Thermoluminescent Dating: Assessment of Alpha Particle Contribution. Archaeometry 17:132-138.

Akazawa, J. (ed.) 1996 Prehistoric Mongolid Dispersals. Oxford University Press, Oxford.

Alexander, H. 1987 Putu: A Fluted Point Site in Alaska. Simon Fraser University Publication No. 17. Archaeology Press, Simon Fraser University, Burnaby.

Alley, R.B., D.A. Meese, C.A. Shuman, A.J. Gow, K.C. Taylor, P.M. Grootes, J.W. White, M. Ram, E.D. Waddington, P.A. Mayewski and G.A. Zielinski 1993 Abrupt Increase in Greenland Snow Accumulation at the End of the Younger Dryas Event. Nature 362:527-529.

Alroy, J. 2001a Did Human Hunting Cause Extinction? Science 294:1459-1462.

2001b A Multispecies Overkill Simulation of the End-Pleistocene Megafaunal Mass Extinction. Science 292:1893-1896.

Anderson, D.D. 1968 A Stone Age Campsite at the Gateway to America. Scientific American 218(6): 24-33.

1988 Onion Portage: The Archaeology of a Stratified Site from the Kobuk River, Northwest Alaska. Anthropological Papers of the University of Alaska 22(1-2).

Anderson, D.G. and C. Gillam 2000 Paleoindian Colonization of the Americas: Implications from an Examination of Physiography, Demography and Artifact Distribution. American Antiquity 65(1):43-66.

193

Anderson, L.L., F.S. Hu, D.M. Nelson, R.J. Petit and K.N. Paige 2006 Ice-Age Endurance: DNA Evidence of a White Spruce Refugium in Alaska. Proceedings of the National Academy of Sciences 103(33):12447-12450.

Andrefsky, W., Jr. 1994 Raw Material Availability and the Organization of Technology. American Antiquity 59(1):21-34

1998 Lithics: Macroscopic Approaches to Analysis. Cambridge Manuals in Archaeology. Cambridge University Press, Cambridge.

Bacon, G.H. 1983 Talkeetna-Lower Susitna River Basin. In Susitna River Basin Study, Cultural Resource Assessment of Willow, Talkeetna and Beluga Areas. pp. 1-117 edited by D. Reger. U.S. Department of Agriculture and the State of Alaska, Department of Natural Resources Report, Anchorage.

Bacon, G.H. and C.E. Holmes 1980 Archaeological Survey and Inventory of Cultural Resources at Fort Greely, Alaska, 1979. Final Report to U.S. Corp of Engineers Alaska District, Alaskarctic, Fairbanks.

Bacon, G.H., J. Kari and T.M. Cole 1983 Cultural Resources Assessment: Talkeetna-Lower Susitna River Basin, Southcentral Alaska. Report to U.S. Department of Agriculture, Soil Conservation Service. Alaskarctic, Anchorage.

Bamforth, D. and P. Bleed 1997 Technology, Flaked Stone Technology and Risk. In Rediscovering Darwin Evolutionary Theory and Archaeological Explanation, edited by C. Barton and G. Clark, pp. 109-129. Archaeological Papers of the American Anthropological Association No. 7.

Barton, C., G. Clark, D.R. Yesner and G. Pearson (eds.) 2004 The Settlement of the American Continents: A Multidisciplinary Approach to Human Biogeography. University of Arizona Press, Tucson.

Beaton, J.M. 1991 Colonizing Continents: Some Problems from Australia and the Americas. In The First Americans: Search and Research, edited by T. Dillehay and D. Meltzer, pp. 209-230. CRC Press, Boca Raton.

Begét, J.E. 1983 Radiocarbon-dated Evidence of Worldwide early Holocene Climate Change. Geology 11:389-393. 194

1990 Middle Wisconsin Climate Fluctuations Recorded in Central Alaskan Loess. Géographie Physique et Quaternaire 44:3-13.

Begét, J.E., R.D. Reger, D. Pinney, T. Gillispie and K. Campbell 1991 Correlation of the Holocene Jarvis Creek, Tangle Lakes, Cantwell and Hayes Tephras in Southcentral and Central Alaska. Quaternary Research 35:174-189.

Benedict, J. 1996 The Game Drives of Rocky Mountain National Park. Research Report No 7. Center for Mountain Archaeology, Ward.

Betts, R.C. 1987 Archaeological Investigations at Butte Lake, Alaska: A Report to the University of Alaska Museum Geist Fund. Report on file at the Alaska Office of History and Archaeology, Anchorage.

Bever, M. 2001a Stone Tool Technology and the Mesa Complex: Developing a Framework of Alaskan Paleoindian Prehistory. Arctic Anthropology 38(2):98-118.

2001b An Overview of Alaskan Late Pleistocene Archaeology: Historical Themes and Current Perspectives. Journal of World Prehistory 15:125-191.

2006 Too Little Too Late? The Radiocarbon Chronology of Alaska and the Peopling of the New World. American Antiquity (71): 595-620.

2008 Distinguishing Holocene Microblades from a Paleoindian Component at the Mesa Site, Alaska. Journal of Field Archaeology 33(2):133-150.

Bigelow, N.H. 1997 Late Quaternary Vegetation and Lake Level Changes in Central Alaska. Unpublished Ph.D. dissertation, Department of Anthropology, University of Alaska, Fairbanks.

Bigelow, N.H. and M.E. Edwards 2001 A 14,000 yr Paleoenvironmental Record from Windmill Lake, Central Alaska: Lateglacial and Holocene Vegetation in the Alaska Range. Quaternary Science Reviews 20:203-215.

Bigelow, N.H. and W.R. Powers 2001 Climate, Vegetation and Archaeology 14,000-9000 cal yr B.P. in Central Alaska. Arctic Anthropology 38(2):171-195.

195

Binford, L.R. 1978a Dimensional Analysis of Behavior and Site Structure: Learning from an Eskimo Hunting Stand. American Antiquity 43:330–361.

1978b Nunamiut Ethnoarchaeology. Academic Press, New York

1980 Willow Smoke and Dogs’ Tails: Hunter-Gatherer Settlement Systems and Archaeological Site Formation. American Antiquity 45(1):1-17.

1983 In Pursuit of the Past: Decoding the Archaeological Record. Thames and Hudson, New York.

Björck, S. 1996 Late Weichselian/Early Preboreal Development of the Öresund Strait; a Key Area for Northerly Mammal Immigration. In The Earliest Settlement of Scandinavia and its Relationship with Neighboring Areas, edited by L. Larsson, pp. 123-134. Acta Archaeologica Lundensia, No. 24, Almquist and Wiksell International, Stockholm.

Björck, S., M. J. C. Walker, L. C. Cwynar, S. Johnsen, K. Knudsen, J. J. Lowe, B. Wohlfarth and INTIMATE Members 1998 An Event Stratigraphy for the Last Termination in the North Atlantic Region Based on the Greenland Ice Core Record: A Proposal by the INTIMATE Group. Journal of Quaternary Science 13:283-292.

Blades, B.S. 1999 Aurignacian Lithic Economy and Early Modern Human Mobility: New Perspectives from Classic Sites in the Vezere Valley of France. Journal of Human Evolution 37(1): 91-120.

2001 Aurignacian Lithic Economy: Ecological Perspectives from Southwestern France. Plenum Publishers, New York.

2003 End Scraper Reduction and Hunter-Gatherer Mobility. American Antiquity 68(1):141-156.

Bleed, P. 1986 The Optimal Design of Hunting Weapons: Maintainability or Reliability. American Antiquity 51(4):737-747.

Bonnichsen, R. and K.L. Turnmire (eds.) 1999 Ice Age Peoples of North America: Environments, Origins and Adaptations of the First Americans. Center for the Study of the First Americans, Oregon State University Press, Corvallis.

196

Bowers, P.M. 1978 Research Summary: 1977 Investigations of the Carlo Creek Archaeological Site, Central Alaska. University of Alaska Museum, Fairbanks.

1979 The Cantwell Ash Bed, A Holocene Tephra in the Central Alaska Range. Alaska Division of Geological & Geophysical Surveys 61.

1980 The Carlo Creek Site: Geology and Archaeology of an Early Holocene Site in the Central Alaska Range. Anthropology and Historic Preservation Cooperative Park Studies Unit. University of Alaska, Fairbanks.

Bowers, P.M. and O.K. Mason 1992 Cultural Resources Inventory and Assessment of a Proposed Power Transmission Line between Cantwell and McKinley Village, Alaska. Northern Land Use Research, Fairbanks.

Bowers, P.M., O.K. Mason, S.L. Ludwig, A.S. Higgs and C.W. Smythe 1995 Cultural Resources Inventory and Assessment of the Proposed Healy to Fairbanks Northern Intertie, South Route and Tanana Flats Alternatives. Submitted to Golden Valley Electric Association, Fairbanks. Northern Land Use Research Inc. Report No. 30, Fairbanks.

Bowers, P.M. and J.D. Reuther 2008 AMS Re-dating of the Carlo Creek Site, Nenana Valley, Central Alaska. Current Research in the Pleistocene 25:58-61.

Bradley, B. and D. Stanford 2004 The North Atlantic Ice-edge Corridor: A Possible Paleolithic Route to the New World. World Archaeology 36(4):459-478.

Brantingham, P., S. Kuhn and K. Kerry 2004 The Early Paleolithic Beyond Western Europe. University of California Press, Berkeley.

Brigham-Grette, J. 2001 New Perspectives on Beringian Quaternary Paleogeography, Stratigraphy and Glacial History. Quaternary Science Reviews. 20:15-24.

Briner, J.P. and D.S. Kaufman 2008 Late Pleistocene Mountain Glaciation in Alaska: Key Chronologies. Journal of Quaternary Science 23(6-7):659-670.

197

Brook, B.W. and D. Bowman 2002 Explaining the Pleistocene Megafaunal Extinctions: Models, Chronologies and Assumptions. Proceedings of the National Academy of the Sciences 99:14624- 14627.

2004 The Uncertain Blitzkrieg of Pleistocene Megafauna. Journal of Biogeography 31:517-523.

Brubaker, L.B., P.M. Anderson, M.E. Edwards and A.V. Lozhkin 2005 Beringia as a Glacial Refugium for Boreal Trees and Shrubs: New Perspectives from Mapped Pollen Data. Journal of Biogeography 32:833-848.

Brubaker, L., P.M. Anderson and F.S. Hu 2001 Vegetation Ecotone Dynamics in Southwest Alaska during the Late Quaternary. Quaternary Science Reviews 20:175-188.

Buck, C.E. and E. Bard 2007 Rapid Communication: A Calendar Chronology for Pleistocene Mammoth and Horse Extinction in North America based on Bayesian Radiocarbon Calibration. Quaternary Science Reviews 26: 2031-2035.

Calkin, P.E., G.C. Wiles and D.J. Barclay 2001 Holocene Coastal Glaciation of Alaska. Quaternary Science Reviews 20:449-461.

CalPal 2006 University of Cologne Radiocarbon Calibration and Paleoclimatic Research Package, May edition.

Carlson, R.L. 1991 Clovis from the Perspective of the Ice-Free Corridor. In Clovis Origins & Adaptations, edited by R. Bonnichsen and K.L. Turnmire, pp. 81-90. Center for the Study of the First Americans, Oregon State University, Corvallis.

Child, J.K., J.E. Begét and A. Werner 1998 Three Holocene Tephras Identified in Lacustrine Sediment Cores from the Wonder Lake Area, Denali National Park and Preserve, Alaska, USA. Arctic and Alpine Research 30(1):89-95.

Clark, D.W. 1995 Batza Téna: The Trail to Obsidian. Arctic Anthropology. 32(1):82-91.

1998 Kodiak Island: The Later Cultures. Arctic Anthropology 35(1)172-186.

198

Clark, D.W. and A.M. Clark 1991 Batza Téna: Trail to Obsidian. Archaeological Survey of Canada Mercury Series, 147. Canadian Museum of Civilization, Hull.

Clarkson, C. 2002 An Index of Invasiveness for the Measurement of Unifacial and Bifacial Retouch: A Theoretical, Experimental and Archaeological Verification. Journal of Archaeological Science 29(1): 65-75.

Close, A. 2000 Reconstructing Movement in Prehistory. Journal of Archaeological Method & Theory 7(1): 49-77.

Coffman, S.C. 2006 Tool Stone Procurement in Southcentral Alaska. Paper presented at the 14th Annual Meeting of the Arctic Conference. Eugene, Oregon. October 20, 2006.

Collins, M. 1999 Clovis Blade Technology: A Comparative Study of the Keven Davis Cache, Texas. Texas Archaeology and Ethnology Series, edited by Thomas R. Hester, University of Texas Press, Austin.

Cook, J. 1969 Early Prehistory of Healy Lake, Alaska. Unpublished Ph.D. dissertation, Department of Anthropology, University of Wisconsin, Madison.

1975 Archaeology of Interior Alaska. The Western Canadian Journal of Anthropology. Special Issue: Athapaskan Archaeology (5): 3-4.

1995 Characterization and Distribution of Obsidian in Alaska. Arctic Anthropology (32)1:92-100.

1996 Healy Lake. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 323-328. University of Chicago Press, Chicago.

Coutler, H.W., D.M. Hopkins, T. Karstrom, T. Péwé and C. Wahrhaftig 1965 Extent of Glaciation in Alaska Map I-415. US Geological Survey, Washington D.C. Miscellaneous Geologic Investigations.

Craig, N., J. Speakman, R.S. Popelka-Filcoff, M.D. Glascock, J.D. Robertson, M.S. Shackley, M.S. Aldenderfer 2007 Comparison of XRF and PXRF for Analysis of Archaeological Obsidian from Southern Peru. Journal of Archaeological Science 34:2012-2024.

199

Davis, S.D. 1989 The Hidden Falls Site, Baranof Island, Alaska. Aurora. Monograph Series 4. Alaska Anthropological Association, Fairbanks.

DeLuca, T., O. Zackrisson, M.C. Nilsson and A. Sellstedt 2002 Quantifying Nitrogen-Fixation in Feather Moss Carpets of Boreal Forests. Nature 419:917-920.

Derevianko, A.P. 1996 Late Pleistocene Sites of the Selemdga River Basin. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F. H. West, pp. 282-289. University of Chicago Press, Chicago.

Dikov, N.N. 1996 Kamchatka Peninsula. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 244-250. University of Chicago Press, Chicago.

Dixon, E.J. 1975 The Gallagher Flint Station: An Early Man Site on the North Slope, Arctic Alaska and its Role in Relation to the Bering Land Bridge. Arctic Anthropology 12(1):68- 75.

1985 Cultural Chronology of Central Interior Alaska. Arctic Anthropology 22(1):47-66.

1993 Quest for the Origins of the First Americans. University of New Mexico Press, Albuquerque.

1999 Bones, Boats and Bison: Archaeology and the First Colonization of Western North America. University of New Mexico Press, Albuquerque.

2001 Human Colonization of the Americas: Timing, Technology and Process. Quaternary Science Reviews. 20:277-299.

in press Arrows, Atl Atls and Cultural Historical Conundrums. In From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia, edited by T. Goebel and I. Buvit. Texas A&M University Press, College Station.

Dixon, E.J., W. Manley and C.M. Lee 2005 The Emerging Archaeology of Glaciers and Ice Patches: Examples from Alaska’s Wrangell-St. Elias National Park and Preserve. American Antiquity 70(1):129- 143.

200

Dixon, E.J. and G.S. Smith 1990 A Regional Application of Tephrochronology in Alaska. In Archaeological Geology of North America, edited by N.P Lasca and J. Donahue, pp.383-398. Geological Society of America, Centennial Special Volume 4.

Dixon, E.J., G.S. Smith, W. Andrefsky, Jr., B.M. Saleeby and C.J. Utermohle 1985 Susitna Hydroelectric Project: Cultural Resources Investigations 1979-1985. University of Alaska Museum Research Report, Fairbanks.

Dortch, J.M. 2006 Defining the Timing of Glaciation in the Central Alaska Range. Unpublished Master’s thesis, University of Cincinnati.

Dugmore, A., G. Larsen and A.J. Newton 2004 Tephrochronology and its Application to Late Quaternary Environmental Reconstruction, with Special Reference to the North Atlantic Islands. In Tools for Constructing Chronologies: Crossing Disciplinary Boundaries, edited by C.E. Buck and A. Millard, pp. 173-187. Springer-Verlag, New York.

Dumond, D.E. 1975 Archaeological Research on the Alaska Peninsula. University of Oregon. Submitted to Secretary of the Interior. Report No. 77-AK-048 & 75-AK-051.

1998 Maritime Adaption on the Northern Alaska Peninsula. Arctic Anthropology 35(1)187-203.

2001 The Archaeology of Eastern Beringia: Some Contrasts and Connections. Arctic Anthropology 38(2):196-205.

Dumond, D.E. and R. Bland 1995 Holocene Prehistory of the Northern-most North Pacific. Journal of World Prehistory. 9(4):401-451.

Dumond, D.E. and R. Knecht 2001 An Early Blade Site in the Eastern Aleutians. University of Oregon Anthropological Papers 58: 9-34.

Easton, N.A., G. MacKay, P.B. Young, P. Schnurr and D.R. Yesner 2008 Nenana in Canada—Emergent Evidence of the Pleistocene Transition of Yukon’s Southwest Beringia as Revealed by the Little John Site (KDVO-6) and Related Regional Survey. Paper presented at the 73rd Annual Meetings of the Society for American Archaeology, Vancouver.

201

Edwards, M.E. and E.D. Barker Jr. 1994 Climate and Vegetation in Northeastern Alaska 18,000 yr B.P.-Present. Palaeogeography, Palaeoclimatology, Palaeoecology 109: 127-135.

Edwards, M.E. and P.F. McDowell 1991 Interglacial Deposits at Birch Creek, Northeast Interior Alaska. Quaternary Research 35:41-52.

Edwards, M.E., C.J. Mock, B.P. Finney, V.A. Barber and P.J. Bartlein 2001 Potential Analogues for Paleoclimatic Variations in Eastern Interior Alaska during the past 14,000 yr: Atmospheric-Circulation Controls of Regional Temperature and Moisture Responses. Quaternary Science Reviews 20:189-202.

Eldridge, K.A. 2008 Staying out of Sight: Hunting Blind Sites in Alaska. Paper presented at the 35th Annual Meetings of the Alaskan Anthropological Association, Anchorage.

Elias, S.A. 2001 Beringian Paleoecology: Results from the 1997 Workshop. Quaternary Science Reviews 20:7-13.

Ellis, C.J. 1997 Factors Influencing the Use of Stone Projectile Tips. In Projectile Technology. edited by H. Knecht, pp. 37–74. New York, Plenum.

Elston, R. and P.J. Brantingham 2002 Microlithic Technology in Northern Asia: A Risk-Minimizing Strategy of the Late Paleolithic and Early Holocene. In Thinking Small: Global Perspectives on Microlithization, edited by R. Elston and S. Kuhn, pp 103-116. Anthropological Papers of the American Anthropological Association. No. 12.

Eriksen, B.V. 2002 Reconsidering the Geochronological Framework of Lateglacial Hunter-gatherer Colonization of Southern Scandinavia. In Recent studies in the Final Palaeolithic of the European Plain, edited by B.V. Eriksen and B. Bratlund, pp.25-41. Proceedings of a U.I.S.P.P. Symposium, Stockholm. Jutland Archaeological Society Publications Vol. 39. Højbjerg.

Esdale, J.A. 2009 Lithic Production Sequences and Toolkit Variability: Examples from the Middle Holocene, Northwest Alaska. Unpublished Ph.D. dissertation Department of Anthropology, Brown University, Providence.

202

Feder, K.L. 1980 Waste Not, Want Not-Differential Lithic Utilization and Efficiency. North American Archeologist 2: 193-205.

Ferguson, D.E. 1997 Gallagher Flint Station, Locality I: A Reappraisal of Proposed Late Pleistocene Site in the Sagavanirktok River Valley, Arctic Alaska. Unpublished Master’s thesis, Department of Anthropology, University of Alaska, Fairbanks.

Fiedel, S.J. 1999 Older than We Thought: Implications of Corrected Dates for Paleoindians. American Antiquity 64(1):95-115.

2000 The Peopling of the New World: Present Evidence, New Theories and Future Directions. Journal of Archaeological Research 8(1):39-103.

Flenniken, J. 1987 The Paleolithic Dyuktai Pressure Blades Technique of Siberia. Arctic Anthropology 24(2):117-132.

Forman, S. 2007 Optically Stimulated Luminescence (OSL) Ages and Associated Chronologic Data for Fine-Grained Sediments (4-11 μm) from near Talkeetna, Alaska. Luminescence Dating Research Laboratory Report, Department of Earth and Environmental Sciences, University of Illinois, Chicago.

Frison, G.C. 2004 Survival by Hunting: Prehistoric Human Predators and Animal Prey. University of California Press, Berkeley.

Frison, G.C., R.L. Andrews, J.M. Adovasio, R.C. Carlisle and R. Edgar 1986 A Late Paleoindian Animal Trapping Net from Northern Wyoming. American Antiquity 51(2):352-61.

Frison, G.C., M. Wilson, D.J. Wilson 1976 Fossil Bison and Artifacts from an Early Altithermal Period Arroyo Trap in Wyoming. American Antiquity 41(1):28-57.

Gal, R. 2002 Providence and Frugality: Tools for High Latitude Living. Paper presented at the 29th Annual Meeting of the Alaska Anthropological Association. April 4-6, Anchorage.

203

Gamble, C., W. Davies, P. Pettitt and M. Richards 2004 Climate Change and Evolving Human Diversity in Europe During the Last Glacial. The Royal Society 359(1442): 243-254.

Gillispie, T.E. 1992 Final Report: Archaeological Data Recovery and Evaluation, Landmark Gap Trail, Alaska. GDM, Inc. Manuscript submitted to BLM, Glennallen District Office.

Glasscock, M., J. Speakman and H. Neff 2007 Archaeometry at the University of Missouri Research Reactor and the Provenance of Obsidian in North America. Archaeometry 49(2):343–357.

Goebel, T. 1992 Geoarchaeological Research at Teklanika West, Denali National Park, Alaska. Preliminary Report submitted to the National Park Service, Alaska Support Office, Anchorage.

1996 Addendum Recent Research at Teklanika West: Site Stratigraphy and Dating. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 341-343. University of Chicago Press, Chicago.

1999 Pleistocene Human Colonization of Siberia and Peopling of the Americas: An Ecological Approach. Evolutionary Anthropology 8(6):208-227.

2002 The “Microblade Adaptation” and Re-colonization of Siberia during the Late Upper Pleistocene. In Thinking Small: Global Perspectives in Microlithization, edited by S. Kuhn and R. Elston, pp. 117-131. Archaeological Papers of the American Anthropological Association No. 12. Washington D.C.

2004 The Search for a Clovis Progenitor in Sub-Arctic Siberia. In Entering America: Northeast Asia and Beringia Before the Last Glacial Maximum, edited by David B. Madsen, pp. 311-356. University of Utah Press, Salt Lake City.

Goebel, T. and N.H. Bigelow 1996 Panguingue Creek. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 366-370. University of Chicago Press, Chicago.

Goebel, T. and I. Buvit (eds.) in press From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia. Texas A&M University Press, College Station.

204

Goebel, T., W.R. Powers and N.H. Bigelow 1991 The Nenana Complex of Alaska and Clovis Origins. In Clovis Origins and Adaptations, edited by R. Bonnichsen and K.L. Turnmire, pp. 49-80. Center for the Study of the First Americans, Oregon State University, Corvallis.

Goebel, T., W.R. Powers, N.H. Bigelow and A.S. Higgs 1996 Walker Road. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp 356-363. University of Chicago Press, Chicago.

Goebel, T., M.R. Waters, I. Buvit, M.V. Konstantinov and A.V. Konstantinov 2000 Studenoe-2 and the Origins of Microblade Technologies in the Transbaikal, Siberia. Antiquity 74:567-575.

Goebel, T., M.R. Waters and M. Dikova 2003 The Archaeology of Ushki Lake, Kamchatka and the Pleistocene Peopling of the Americas. Science 301:501-505.

Graf, K.E. 2005 Abandonment of the Siberian Mammoth-Steppe During the LGM: Evidence from the Calibration of 14C-Dated Archaeological Occupations. Current Research in the Pleistocene 22:2-5.

2008 Uncharted Territory: Late Pleistocene Hunter-Gatherer Dispersals in the Siberian Mammoth Steppe. Unpublished PhD dissertation, Department of Anthropology, University of Nevada, Reno.

Greenberg, J.H., C.G. Turner and S.L. Zegura 1986 The Settlement of the Americas: A Comparison of the Linguistic, Dental and Genetic Evidence. Current Anthropology 27:477-97.

Griffin, K. 1990 An Overview and Assessment of Archeological Resources, Denali National Park and Preserve, Alaska. Research/Resources Management Report AR-16, Department of Interior, National Park Service, Alaska Regional Office, Anchorage.

Guthrie, R.D. 1983 Paleoecology of the Site and Its Implications for Early Hunters. In Dry Creek: Archaeology and Paleoecology of a Late Pleistocene Alaskan Hunting Camp, edited by W. R. Powers, R. D. Guthrie and J. F. Hoffecker, pp. 209–287. Report submitted to the U.S. National Park Service, Anchorage.

1990 Frozen Fauna of the Mammoth Steppe: The Story of Blue Babe. University of Chicago Press, Chicago.

205

2001 Origin and Causes of the Mammoth Steppe: A Story of Cloud Cover, Woolly Mammal Tooth Pits, Buckles and Inside-Out Beringia. Quaternary Science Reviews 20:54-574.

2006 New Carbon Dates Link Climatic Change with Human Colonization and Pleistocene Extinctions. Nature 441:207-209.

Hall, E.S., Jr. 1982 The Potential Significance of Small, Single Component Archaeological Sites for Elucidating the Culture History of Northern Alaska. Anthropological Papers of the University of Alaska 20(1-2):7-13.

Hall, R., D. Roy and D. Boling 2004 Pleistocene Migration Routes into the Americas: Human Biological Adaptations and Environmental Constraints. Evolutionary Anthropology 12:132-144.

Hamilton, T.D., T.A. Ager and S.W. Robinson 1983 Late Holocene Ice Wedges near Fairbanks, Alaska U.S.A: Environmental Settings and History of Growth. Arctic and Alpine Research 15:157-168.

Hamilton, T.D. and T. Goebel 1999 Late Pleistocene Peopling of Alaska. In Ice Age Peoples of North America: Environments, Origins and Adaptations of the First Americans, edited by R. Bonnichsen and K. L. Turnmire, pp. 156-199. Center for the Studies of the First Americans, Oregon State University, Corvallis.

Hamilton, T.D. and R.M. Thorson 1983 The Cordilleran Ice Sheet in Alaska. In Late-Quaternary Environments of the United States, edited by S. C. Porter. The Late Pleistocene 1:38-52. University of Minnesota Press, Minneapolis.

Hare, P.G. 1995 Holocene Occupations in the Southern Yukon: New Perspectives from the Annie Lake Site. Occasional Papers in Archaeology Vol. 5, Heritage Branch, Government of the Yukon, Whitehorse.

Hare, P.G., S. Greer, R. Gotthardt, R. Farnell, V. Bowyer and C. Schweger 2004 Ethnographic and Archaeological Investigations of Alpine Ice Patches in Southwestern Yukon, Canada. Arctic 57(3):260-272.

Harris, M. 1968 The Rise of Anthropological Theory. Harper and Row, New York.

Harrison, A.E. 1964 Ice surges on the Muldrow Glacier, Alaska. Journal of Glaciology 5(39):365-368. 206

Haynes, G.A. 1993 Noncultural Modifications to Mammalian Bones in Sites of Mass Deaths and Serial Predation. Anthropologie 29(3):151-156.

2002 The Early Settlement of North America. Cambridge University Press, Cambridge.

Hedges, R.E.M. and G.J. Van Klinken 1992 A Review of Current Approaches in the Pretreatment of Bone for Radiocarbon Dating by AMS. Radiocarbon 34:279-291.

Helwig, K., V. Monahan and J. Poulin 2008 The Identification of Hafting Adhesive on a Slotted Antler Point from a Southwest Yukon Ice Patch. American Antiquity 73(2):279-288.

Hendy, I.L., J.P. Kennett, E.B. Roark and B.L. Ingram 2002 Apparent Synchroneity of Submillennial Scale Climate Events between Greenland and Santa Barbara Basin, California from 30-10 ka. Quaternary Science Reviews 21:1167-1184.

Henn, W. 1978 Archaeology on the Alaska Peninsula: the Ugashik Drainage, 1973-1975. University of Oregon Anthropological Papers No.14, Eugene.

Hoffecker, J.F. 1996 Moose Creek. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp 363-366. University of Chicago Press, Chicago.

2001 Late Pleistocene and Early Holocene Sites in the Nenana River Valley, Central Alaska. Arctic Anthropology 38(2):139-153.

2002 Desolate Landscapes: Ice Age Settlement in Eastern Europe. Rutgers University Press, New Brunswick, New Jersey.

in press Assemblage Variability in Beringia: The Mesa Factor. In From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia, edited by T. Goebel and I. Buvit. Texas A&M University Press, College Station.

Hoffecker, J.F. and S.A. Elias 2003 Environment and Archaeology in Beringia. Evolutionary Anthropology 12:34-49.

2007 Human Ecology of Beringia. Columbia University Press, New York.

207

Hoffecker, J.F. and W.R. Powers 1985 North Alaska Range Early Man Project. National Geographic Society Research Reports No. 19, Washington, DC.

1996 Little Panguingue Creek. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 371-374. University of Chicago Press, Chicago.

Hoffecker, J.F., W.R. Powers and N.H. Bigelow 1996a Dry Creek. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 343-352. University of Chicago Press, Chicago.

Hoffecker, J.F., W.R. Powers and T. Goebel 1993 The Colonization of Beringia and the Peopling of the New World. Science 259:46-53.

Hoffecker, J.F., W.R, Powers and P.G. Phippen 1996b Owl Ridge. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 353-356. University of Chicago Press, Chicago.

Hogan, B. 1974 Two High Altitude Game Trap Sites in Montana. Unpublished Master’s thesis, University of Montana, Missoula.

Holmes, C.E. 1988 An Early Post Paleo-Arctic Site in the Alaska Range. Paper presented at the 15th Annual Meeting of the Alaska Anthropological Association, Fairbanks.

1996 Broken Mammoth. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 312-318. University of Chicago Press, Chicago.

1998 New Data Pertaining to Swan Point, the Oldest Microblade Site Known in Alaska. Current Research in the Pleistocene 15:21-22.

2001 Tanana River Valley Archaeology crica 14,000 to 9,000 B.P. Arctic Anthropology 38(2):154-170. in press The Beringian and Transitional Periods in Alaska: Technology of the East Beringian Tradition as Viewed from Swan Point. In From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia, edited by T. Goebel and I. Buvit. Texas A&M University Press, College Station. 208

Holmes, C.E. and G.H. Bacon 1982 Holocene Bison in Central Alaska: A Possible Explanation for Technological Conservatism. Paper Presented at the 9th Annual Meeting of the Alaska Anthropological Association, Fairbanks. Manuscript on file at the National Park Service, Alaska Support Office, Anchorage.

Holmes, C.E. and B.A. Crass 2003 Early Cultural Components in Central Alaska: An Update from Swan Point. Paper presented at the 30th Annual Meeting of the Alaska Anthropological Association, Fairbanks.

Holmes, C.E., R. VanderHoek and T. Dilley 1996 Swan Point. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 319-322. University of Chicago Press, Chicago.

Hopkins, D.M. 1959 Cenozoic History of the Bering Land Bridge. Science 129(3362):1519-1528.

1967 The Bering Land Bridge. Stanford University Press, Stanford.

Hundertmark, K.J., G.F. Shields, I.G. Udina, R.T. Bowyer, A.A. Danilkin and C.C. Schwartz 2002 Mitochondrial Phylogeography of Moose (Alces alces): Late Pleistocene Divergence and Population Expansion. Molecular Phylogenetics and Evolution. 22(3)375-387.

Jain, M., L. Botter-Jensen and A.K. Singhvi 2003 Dose Evaluation using Multiple-Aliquot Quartz OSL: Test of Methods and a New Protocol for Improved Accuracy and Precision: Radiation Measurements 37:67- 80.

Jordan, J.W. 2001 Late Quaternary Sea Level Change in Southern Beringia: Postglacial Emergence of the Western Alaska Peninsula. Quaternary Science Reviews 20:509-523.

Jordan, R.H. 1992 A Maritime Paleo-Arctic Assemblage from Crag Point, Kodiak Island, Alaska. Anthropological Papers, University of Alaska 24:127-140.

Kari, J. and J. Fall 2003 Shem Pete's Alaska: The Territory of the Upper Cook Inlet Dena'ina. University of Alaska Press, Fairbanks.

209

Kaufman, D.S., T.A. Ager, N.J. Anderson, P.M. Anderson, J.T. Andrews, P.J. Bartlein, L.B. Brubaker, L.L. Coats, L.C. Cwynar, M.L. Duvall, A.S. Dyke, M.E. Edwards, W.R. Eisner, K. Gajewski, A. Geirsdottir, F.S. Hu, A.E. Jennings, M.R. Kaplan, M.W. Kerwin, A.V. Lozhkin, G.M. MacDonald, G.H. Miller, C.J. Mock, W.W. Oswald, B.L. Otto- Bliesner, D.F. Porinchu, K. Ruhland, J.P. Smol, E.J. Steig and B.B. Wolf 2004 Holocene Thermal Maximum in the Western Arctic (0-180° W). Quaternary Science Reviews 23:529-560.

Kaufman, D.S. and W.F. Manley 2004 Pleistocene Maximum and Late Wisonsinan Glacier Extents across Alaska, U.S.A. In Quaternary Glaciations: Extent and Chronology, Part II: North America, edited by J. Ehlers and P.L. Gibbard, pp. 9-27. Elsevier, Amsterdam.

Kedrowski, B.L., B.A. Crass, J.A. Behm, J.C. Luetke, A.L. Nichols, A.M. Moreck and C.E. Holmes 2008 GC/MS Analysis of Fatty Acids from Ancient Hearth Residues at the Swan Point Archaeological Site. Archaeometry 51(1):110-122.

Kehoe, T.F. and A.B. Kehoe 1960 Observations on the Butchering Technique at a Prehistoric Bison-Kill in Montana. American Antiquity (25)3:420-423.

Kelly, R. 1983 Hunter Gatherer Mobility Strategies. Journal of Anthropological Research 39:277-306.

1992 Mobility/Sedentism: Concepts, Archaeological Measures and Effects. Annual Review of Anthropology 21(1): 43-66.

1995 The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways. Smithsonian Institution Press, Washington D.C.

2003 Colonization of New Land by Hunter-Gatherers: Expectations and Implications Based on Ethnographic Data. In Colonization of Unfamiliar Landscapes: The Archaeology of Adaptation, edited by M. Rockman and J. Steele, pp. 44-58. Routledge, London.

Kelly, R. and L.C. Todd 1988 Coming into the Country: Early Paleoindian Hunting and Mobility. American Antiquity 53(2):231-244.

210

Kenyon, D. 1997 Large Kill Sites and the Potential for Illuminating Provisioning Behavior: Preliminary Thoughts and Expectations. In Caribou and Reindeer Hunters of the Northern Hemisphere, edited by, L. Jackson and P. Thacker, pp. 1-26. Worldwide Archaeology Series, Avebury, Aldershot.

Keyeser, J. D. 1974 The LaMarche Game Trap: An Early Historic Game Trap in Southwestern Montana. Plains Anthropologist 19(55):173-79.

Klein, R. 2000 Archaeology and the Evolution of Human Behavior. Evolutionary Anthropology 9:17-36.

Knecht, R. and R. Davis 2001 A Prehistoric Sequence for the Eastern Aleutians. University of Oregon Anthropological Papers 58:269-288.

Kobayashi, T. 1970 Microblade Industries in the Japanese Archipelago. Arctic Anthropology 7(2):38- 50.

Koch, P.L. and A.D. Barnosky 2006 Late Quaternary Extinctions: State of the Debate. Annual Review of Ecology, Systematics 36:215-250.

Koch, P.L., N.S. Diffenbaugh and K.A. Hoppe 2004 The Effects of Pleistocene Climate and pCO2 Change on C4 Plant Abundance in the South-Central United States. Paleogeography, Paleoclimate, Paleoecology 207:331-357.

Koch, P.L., K.A. Hoppe and S.D. Webb 1998 The Isotopic Ecology of Late Pleistocene Mammals in North America-Part I. Florida. Chemical Geology 152:119-138.

Krasinski, K.E. 2005 Intrasite Spatial Analysis of Late Pleistocene/Early Holocene Archaeological Material from the Broken Mammoth Site. Unpublished Master’s thesis, Department of Anthropology, University of Alaska, Anchorage.

Krasinski, K.E. and D.R. Yesner 2008 Late Pleistocene/Early Holocene Site Structure in Beringia: A Case Study from the Broken Mammoth Site, Interior Alaska. Alaska Journal of Anthropology 6(1/2):27-41.

211

Kruskal, W.H. and W.A. Wallis 1952 Use of Ranks in One-criterion Variance Analysis. Journal of the American Statistical Association 47(260): 583–621.

Kuhn, S. 1991 Unpacking Reduction: Lithic Raw Material Economy in the Mousterian of West- central Italy. Journal of Anthropological Archaeology 10:76–106.

1995 Mousterian Lithic Technology: An Ecological Perspective. Princeton University Press, Princeton.

Kuhn, S.L. and R.G. Elston 2002 Introduction Thinking Small Globally. In Thinking Small: Global Perspectives on Microlithization, edited by R.G. Elston and S.L. Kuhn, pp. 1-7. Archeological Papers of the American Anthropological Association 12.

Kunz, M., M. Bever and C. Adkins 2003 The Mesa Site: Paleoindians Above the Arctic Circle. US Department of the Interior, Bureau of Land Management Alaska Open File Report No. 86. Anchorage.

Kunz, M. and R. Reanier 1994 Paleoindians in Beringia: Evidence from Arctic Alaska. Science 263:660-662.

1995 The Mesa Site: A Paleoindian Hunting Lookout in Arctic Alaska. Arctic Anthropology 32(1:)5-30.

1996 Mesa Site, Iteriak Creek. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 497-504. University of Chicago Press, Chicago.

Kurtén, B. and E. Anderson 1980 Pleistocene Mammals of North America. Columbia University Press, New York.

Kutzbach, J., R. Gallimore, S. Harrison, R. Behling, R. Selin and F. Laarif 1998 Climate and Biomes Simulations for the Past 21,000 Years. Quaternary Science Reviews 17:473-506.

Kuzmin, Ya. V 2008 Temporal Patterns of Existence and Extinction for Wooly Mammoth (Mammuthus primigenius Blum.) in Northern Asia: The 2007 State of the Art. Current Research in the Pleistocene. 25:177-179.

212

Kuzmin, Ya. V. and L. A. Orlova 1998 Radiocarbon Chronology of the Siberian Paleolithic. Journal of World Prehistory. 12(1):1-53.

Kuzmin, Ya. V. and K.B. Tankersley 1996 The Colonization of Eastern Siberia: an Evaluation of the Paleolithic Age Radiocarbon Dates. Journal of Archaeological Science 23 p 577-585.

Larichev, V. 1992 The Upper Paleolithic of Northern Asia: Achievements, Problems and Perspectives. III. Northeastern Siberia and the Russian Far East. Journal of World Prehistory 6 (4):441 -476.

Larichev, V., U. Khol’ushkin, and I. Laricheva 1988 The Upper Paleolithic of Northern Asia: Achievements, Problems and Perspectives I. Western Siberia. Journal of World Prehistory 2(4):359-396.

Larsen, H. 1968 Trail Creek: Final Report on the Excavation of Two Caves on Seward Peninsula, Alaska. Acta Arctica 15.

Laughlin, W. 1975 Aleuts: Ecosystems, Holocene History and Siberian Origins. Science 189:507- 515.

Lemke, K.J. 2000 Holocene Tephrostratigraphy, Southern Kenai Peninsula, Lower Cook Inlet, Alaska. Unpublished Mater thesis, Utah State University, Logan.

Lilliefors, H.W. 1967 On the Kolmogorov-Smirnov Test for Normality with Mean and Variance Unknown. Journal of the American Statistical Association 62:399-402.

Lively, R.A. 1996 Chugwater. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 308-311. University of Chicago Press, Chicago.

Loehr, J., K. Worley, A. Grapputo, J. Carey, A. Veitchs, and D. W. Coltman 2006 Evidence for Cryptic Glacial Refugia from North American Mountain Sheep Mitochondrial DNA. European Society for Evolutionary Biology 19:419-430.

213

Lynch, A.J. 1996 Archeological Investigations of Five Remote Tracts of Land within Denali National Park and Preserve, Alaska 1988, 1989. Department of the Interior, National Park Serve, Alaska System Support Office Division of Cultural Resources, Anchorage.

Lynch, A.J., J.S. Clark, N.H. Bigelow and M.E. Edwards 2002 Geographic and Temporal Variations in Fire History in Boreal Forest Ecosystems of Alaska. Journal of Geophysical Research. 108(D1-8152):1-17.

MacDonald, D. H. 1998 Subsistence, Sex and Cultural Transmission in Folsom Culture. Journal of Anthropological Archaeology 17:217-239.

2004 Early Paleoindians as Estate Settlers: Archaeological, Ethnographic and Evolutionary Insights into the Peopling of the New World. In The Settlement of the American Continents: A Multidisciplinary Approach to Human Biogeography, edited by C.M. Barton, G. A. Clark, D. R. Yesner and G.A. Pearson, pp. 173-176. University of Arizona Press, Tucson.

MacPhee, R.D.E., A.N. Tikhonov, D. Mol, C. de Marliave, H. van der Plicht, A.D. Greenwood, C. Flemming and L. Agenbroad 2002 Radiocarbon Chronologies and Extinction Dynamics of the Late Quaternary Mammalian Megafauna of the Taimyr Peninsula, Russian Federation. Journal of Archaeological Science 29:1017-1042.

Madsen, D.B. (ed.) 2004 Entering the Americas: Northeast Asia and Beringia Before the Last Glacial Maximum. University of Utah Press, Salt Lake City.

Mandryk, C.A., J. Heiner, D. Fedje and R.W. Mathews 2001 Late Quaternary Paleoenvironments of Northwestern North America: Implications for Inland Versus Coastal Migration Routes. Quaternary Science Reviews 20(1-3):301-314.

Mangerud, J., S.T. Andersen, B.E. Berglund and J.J. Donner 1974 Quaternary Stratigraphy of Norden: A Proposal for Terminology and Classification. Boreas 4:109-128.

Mangerund, J and B.E. Berglund 1978 The Subdivision of the Quaternary of Norden: A Discussion. Boreas 7:179–181.

Manley, W.F. and D.S. Kaufman 2002 Postglacial Flooding of the Bering Land Bridge: A Geospatial Animation: INSTAAR, University of Colorado, v1. http://instaar.colorado.edu/QGISL/bering_land_bridge. 214

Manley, W.F., D.S. Kaufman, and J.P. Briner 2001 Pleistocene Glacial History of the Southern Ahklun Mountains, Southwestern Alaska: Soil Development, Morphometric and Radiocarbon Constraints. Quaternary Science Reviews 20:353-370.

Mann, D.H., A. Crowell, T.D. Hamilton and B. Finney 1998 Holocene Geologic and Climatic History Around the Gulf of Alaska. Arctic Anthropology 35(1): 112-131.

Mann, D.H. and T.D. Hamilton 1995 Late Pleistocene and Holocene Paleoenvironments of the North Pacific Coast. Quaternary Science Reviews 14:449-471.

Mann, D.H. and D.M. Peteet 1994 Extent and Timing of the Last Glacial Maximum in Southwestern Alaska. Quaternary Research 42(2):136-148.

Mann, D.H., D. Peteet, R. Reanier and M. Kunz 2002 Responses of an Arctic Landscape to Lateglacial and early Holocene Climatic Changes: The Importance of Moisture. Quaternary Research 21:997-1021

Mann, D.H., R. Reanier, D. Peteet, M. Kunz and M. Johnson 2001 Environmental Change and Arctic Paleoindians. Arctic Anthropology 38(2):119- 138.

Mann, H.B. and D.R. Whitney 1947 On a Test of Whether One of Two Random Variables is Stochastically Larger than the Other. Annals of Mathematical Statistics 18: 50–60.

Martin, P. 1984 Prehistoric Overkill: The Global Model. In Quaternary Extinctions: A Prehistoric Revolution. edited by P.S. Martin and R.G. Klein, pp. 354-403. University of Arizona Press, Tucson.

Mason, O.K., P.M. Bowers and D.M. Hopkins 2001 The Early Holocene Milankovitch Thermal Maximum and Humans: Adverse Conditions for the Denali Complex of Eastern Beringia. Quaternary Science Reviews 20(1-3):525-548.

Matson, R.G. and G.G. Coupland 1995 The Initial Colonization of the Northwest Coast. In Prehistory of the Northwest Coast, edited by R. Madsen and G. Coupland, pp. 49-94. Academic Press, San Diego.

215

Mead, J.I. and D.J. Meltzer 1984 North American late Quaternary Extinctions and the Radiocarbon Record. In Quaternary Extinctions: A Prehistoric Revolution, edited by P.S. Martin and R.G. Klein, pp.440-450, University of Arizona Press, Tuscon.

Medvedev, G.I. and B.P. Merbs 1964 The Place of the Culture of Verkholenskaia Gora in the Archaeological Sequence of the Baikal Region. American Antiquity 29(4): 461-466.

Meltzer, D.J. 2002 What Do You Do When No One’s Been There Before? Thoughts on The Exploration and Colonization of New Lands. In The First Americas: The Pleistocene Colonization of the New World, edited by N. G. Jablonski, pp. 27-58. Memoirs of the California Academy of Sciences, No. 27, San Francisco.

2003 Lessons in Landscape Learning. In Colonization of Unfamiliar Landscapes: The Archaeology of Adaptation, ed. by M. Rockman and J. Steele, pp. 222-241. Routledge, New York.

2004 Modeling the Initial Colonization of the Americas: Issues of Scale, Demography, and Landscape Learning. In The Settlement of the American Continents: A Multidisciplinary Approach to Human Biogeography, ed. by C. M. Barton, G. A. Clark, D. R. Yesner and G. A. Pearson, pp. 123-137. University of Arizona Press, Tucson.

Midwest Geosciences Group n.d. Field Guide for Soil and Stratigraphic Analysis. Available online at: http://www.midwestgeo.com/soilanalysis.htm.

n.d. US Department of Agriculture Soil Texturing Field Flow Chart. Available online at: http://www.midwestgeo.com/soilflowchart.htm.

Mills, R.O. 1994 Radiocarbon calibration of archeological sites from the central Gulf of Alaska. Arctic Anthropology 3(1):126-149.

Mobley, C.M. 1982 The Landmark Gap Trail Site, Tangle Lakes, Alaska: Another Perspective on the Amphitheater Mountain Complex. Arctic Anthropology 19(1):81-102.

1991 The Campus Site. University of Alaska Press, Fairbanks.

1996 The Campus Site. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 296-301. University of Chicago Press, Chicago. 216

Mobley, C.M. and A.J. Morris 1981 Archaeological Investigations at Tangle Lakes, Alaska – 1980. Bureau of Land Management, Anchorage District Office.

Morgan, H.M. 1965 An Archaeological Survey of Mount Mckinley National Park. U.S. Department of Interior, National Park Service Contract 14-10-0434-951.

Morlan, R.E. 1976 Technological Characteristics of Some Wedge-Shaped Cores in North America and northeast Asia. Asian Perspectives 19(1):96-106.

2007 Canadian Archaeological Radiocarbon Database, Canadian Museum of Civilization, Gatineau. Electronic document, http://www.canadianArchaeology.ca/ accessed March 21, 2007.

Mosimann, J.E. and P.S. Martin 1975 Simulating Overkill by Paleoindians. American Scientist 63:304-313.

Moss, M. 1998 Northern Northwest Coast Regional Overview. Arctic Anthropology. 35(1):88- 111.

Muhs, D.R., T.A. Ager, E. Arthur Betis III, J. McGeehin, J.M. Been, J.E. Begét, M.J. Pavich, T.W. Stafford Jr. and D.A. Stevens 2003 Stratigraphy and Paleoclimatic Significance of Late Quaternary Loess-Palaeosol Sequencess of the Last Interglacial-Glacial Cycle in Central Alaska. Quaternary Science Reviews 22:1947-1986.

Murray, A.S. and A.G. Wintle 2003 The Single Aliquot Regenerative Dose Protocol: Potential for Improvements in Reliability. Radiation Measurements 37:377-381.

Neff, J.C., A.R. Townsend, G. Gleixner, S.J. Lehman, J. Turnbull and W.D. Bowman 2002 Variable Effects of Nitrogen Additions on the Stability and Turnover of Soil Carbon. Nature 419:915-917.

Nelson, N.C. 1937 Notes on Cultural Relations between Asia and America. American Antiquity 2(4):267-276.

Newman, J.R. 1994 The Effects of Distance on Lithic Material Reduction Technology. Journal of Field Archaeology 21(4):491-501.

217

Odell, G.H. 1996 Stone Tools: Theoretical Insights into Human Prehistory. Plenum Press, New York.

2004 Lithic Analysis. Springer-Verlag, New York.

Odess, D. and J.T. Rasic 2007 Toolkit Composition and Assemblage Variability: Implications from the Nogahabara I Site, Northern Alaska. American Antiquity 72(4):691-717.

Oswalt, W. 1976 An Anthropological Analysis of Food-Getting Technology. Wiley and Sons, New York.

Owen, L.R. 1988 Blade and Microblade Technology. BAR International Series 441. Oxford.

Pearson, G.A. 1999 Early Occupations and Cultural Sequence at Moose Creek: A Late Pleistocene Site in Central Alaska. Arctic 52:332-345.

Peteet, D.M. and D.H. Mann 1994 Late-Glacial Vegetational, Tephra, and Climatic History of Southwestern Kodiak Island, Alaska. Ecoscience 1:255-267.

Pettitt, P.B. 2005 The Rise of Modern Humans. In The Human Past: World Prehistory and the Development of Human Societies, edited by Chris Scarre, pp. 124-173. Thames and Hudson, London.

Pettitt, P.B., W. Davies, C.S. Gamble and M.B. Richards 2003 Paleolithic Radiocarbon Chronology: Quantifying Our Confidence Beyond Two Half-Lives. Journal of Archaeological Science 30:1685-1693.

Péwé, T.L. 1951 An Observation on Wind-Blown Silt. Journal of Geology 59:399-401.

1975 Quaternary Geology of Alaska. U.S. Geological Survey Professional Paper 835:145.

Péwé, T.L. and R.D. Reger 1983 Middle Tanana River Valley. In Guidebook to Permafrost and Quaternary Geology along the Richardson and Glenn Highways between Fairbanks and Anchorage, Alaska. edited by T. Péwé and R. Reger. Department of Natural Resources, Division of Geological and Geophysical Surveys, Fairbanks. 218

Phippen, P.G. 1988 Archaeology at Owl Ridge: A Pleistocene-Holocene Boundary Aged Site in Central Alaska. Unpublished Master’s thesis, Department of Anthropology, University of Alaska, Fairbanks.

Plaskett, D.C. 1976 Nenana and Teklanika Rivers of Central Alaska-A Cultural Resources Survey. State of Alaska Department of National Resources Division of Parks. Contract No. 02-76-00035, Anchorage.

Potter, B. 2005 Site Structure and Organization in Central Alaska: Archaeological Investigations at Gerstle River. Unpublished Ph.D. dissertation, Department of Anthropology, University of Alaska, Fairbanks.

2008a Exploratory Models of Intersite Variability in Mid-Late Holocene Central Alaska. Arctic 61(4):407-425.

2008b A First Approximation of Holocene Inter-assemblage Variability in Central Alaska. Arctic Anthropology 45(2).

2008c Radiocarbon Chronology of Central Alaska: Technological Continuity and Economic Change. Radiocarbon 50(2):181-204.

Powers, W.R. 1973 Paleolithic Man in Northeast Asia. Arctic Anthropology 10(2):1-106.

Powers, W.R. and R.D. Guthrie 1977 Early Man Studies in Alaska. Proposal submitted to the National Park Service.

Powers, W.R., R.D. Guthrie and J.F. Hoffecker (eds.) 1983 Dry Creek: Archaeology and Paleoecology of a Late Pleistocene Alaskan Hunting Camp. National Park Service Report, Alaska Support Office, Anchorage.

Powers, W.R. and T.D. Hamilton 1978 Dry Creek: A Late Pleistocene Human Occupation in Central Alaska. In Early Man in America, edited by A.L. Bryan, pp. 72-77. Occasional Paper No. 1. Department of Anthropology, University of Alberta, Edmonton.

Powers, W.R. and J.F. Hoffecker 1989 Late Pleistocene Settlement in the Nenana Valley, Central Alaska. American Antiquity 54(2):263-287.

219

Prescott, J.R. and J.T. Hutton 1994 Cosmic Ray Contributions to Dose Rates for Luminescence and ESR Dating: Large Depths and Long-term Time Variations. Radiation Measurements 23:497- 500.

Rainey, F.G. 1940 Archaeological Investigations in Central Alaska. American Antiquity 5(4):299- 308.

Rapp, Jr., G. and C.L. Hill 1998 Geoarchaeology: The Earth-Science Approach to Archaeological Interpretation. Yale University Press, New Haven.

Rasic, J.T. 2003 Ancient Hunters of the Western Brooks Range: Integrating Research and Cultural Resource Management. Alaska Park Science, Winter 21-25. U.S. Department of the Interior, National Park Service, Alaska Support Office, Anchorage.

2008 Paleoalaskan Adaptive Strategies Viewed from Northwestern Alaska. Unpublished Ph.D. dissertation Department of Anthropology, Washington State University, Pullman.

Rasic, J.T. and R. Gal 2000 An Early Lithic Assemblage from the Tuluaq Site, Northwest Alaska. Current Research in the Pleistocene 17:66-68.

Rausch, R. 1951 Notes on the Nunamuit Eskimo and Mammals of the Anaktuvuk Pass Region, Brooks Range, Alaska. Arctic 4:146-195.

Reanier, R.E. 1995 The Antiquity of Paleoindian Materials in Northern Alaska. Arctic Anthropology 32(1):31-50.

Reger, D.R. 1981 A Model for Culture History in Upper Cook Inlet, Alaska. Unpublished Ph.D. Dissertation, Washington State University, Pullman.

1996 Beluga Point. In American Beginnings: The Prehistory and Palaeoecology of Beringia, edited by F.H. West, pp. 433-436. University of Chicago Press, Chicago.

1998 Archaeology of the Northern Kenai Peninsula and Upper Cook Inlet. Arctic Anthropology 35(1):160-171.

220

Reger, D.R. and G. Bacon 1996 Long Lake. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 436-438. University of Chicago Press, Chicago.

Reger, R.D., R.A. Combellick and J. Brigham-Grette 1995 Late-Wisconsin Events in the Upper Cook Inlet Region, Southcentral Alaska. In Short Notes on Alaskan Geology. edited by R. Combellick and F. Tannian. pp. 33- 45. Alaska Division of Geological and Geophysical Surveys Professional Report 117.

Reger, R.D., D.S. Pinney, R.M. Burke and M.A. Wiltse 1996 Catalog and Initial Analyses of Geologic Data Related to Middle to Late Quaternary Deposits, Cook Inlet Region, Alaska. State of Alaska Division of Geological and Geophysical Surveys Report of Investigations 95-6.

Reger, R.D., A.G. Sturmann, E.E. Berg and P.A. Burns 2007 A Guide to the Late Quaternary History of Northern and Western Kenai Peninsula, Alaska. Guidebook 8, State of Alaska, Department of National Resources, Division of Geological and Geophysical Surveys, Anchorage.

Reimer, P.J., G.L. Baillie, E. Bard, A. Bayliss, J.W. Beck, C. Bertrand, P.G. Blackwell, C.E. Buck, G. Burr, K.B. Cutler, P.E. Damon, R.L. Edwards, R.G. Fairbanks, M. Friedrich, T.P. Guilderson, K.A. Hughen, B. Kromer, F.G. McCormac, S. Manning, C. B. Ramsey, R.W. Reimer, S. Remmele, J.R. Southon, M. Stuiver, S. Talamo, F.W. Taylor, J. van der Plicht and C.E. Weyhenmeyer 2004 IntCal04 Terrestrial Radiocarbon Age Calibration, 26-0 ka BP. Radiocarbon 46:1029-1058.

Rhode, D. 1988 Measurement of Archaeological Diversity and the Sample-Size Effect. American Antiquity 53(4):708-716.

Rick, J.W. 1987 Dates as Data: An Examination of the Peruvian Preceramic Radiocarbon Record. American Antiquity 52(1):55-73.

Riehle, J.R. 1994 Heterogeneity, Correlatives and Proposed Stratigraphic Nomenclature of Hayes Tephra Set H, Alaska. Quaternary Research 41:285-288.

Riehle, J.R., P.M. Bowers and T.A. Ager 1990 The Hayes Tephra Deposits, an Upper Holocene Marker Horizon in Southcentral Alaska. Quaternary Research 33:276-290.

221

Riehle, J.R., R.B. Waitt, C.E. Meyer and L.C. Calk 1998 Age of Formation of Kaguyak Caldera, Eastern Aleutian Arc, Alaska, Estimated Tephrochronology. In Geological Studies in Alaska by the U.S. Geological Survey, 1996, edited by J. E. Gray and J. R. Riehle, U.S. Geological Survey Professional Paper 1595, Washington D.C.

Robinson, B., F.H. West and D.R. Reger 1996 Long Lake. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F. H. West. pp. 438-443. University of Chicago Press, Chicago.

Roosevelt, A.C. 2000 Who's on First? There's Still No End to the Controversy Over When and How Humans Populated the New World: A Review of T. Dillehay's The Settlement of the Americas and of E.J. Dixon's Bones, Boats and Bison. Natural History 109(6)76-79.

Ruhe, R.V. 1983 Depositional Environment of late Wisconsin Loess in the Midcontinental United States. In Late-Quaternary Environments of the United States. Vol 1, The Late Pleistocene, edited by H.E. Wright, Jr. and S.C. Porter. pp. 130-137. University of Minnesota Press, Minneapolis.

Saleeby, B.M. 2000 CRMIM: The Quest for Gold. National Park Service-Alaska Region. Research Resources Management Report AR-9, Anchorage.

Sattler, R.A. 1997 Large Mammals in Lower Rampart Cave 1, Alaska: Interspecific Utilization of an eastern Beringian Cave. Geoarchaeology 12(6): 657-688.

Schaaf, J. 1988 The Bering Land Bridge National Preserve: An Archeological Survey. National Park Service Research/Resources Management Report AR-14, Anchorage.

Schiff, C.J., D.S. Kaufman, K.L. Wallace, A. Werner, T. Ku and T.A. Brown 2008 Modeled Tephra Ages from Lake Sediments, Base of Redoubt Volcano, Alaska. Quaternary Geochronology. 3:56-67.

Schmoll, H.R. and L.A. Yehle 1986 Pleistocene Glaciation of the Upper Cook Inlet Basin. In Glaciation in Alaska: The Geologic Record, edited by Hamilton, T.D., K.M. Reed and R.M. Thorson. Alaska Geologic Society, Anchorage.

222

Schurr, T.G. 2004 Molecular Genetic Diversity in Siberians and Native Americans Suggests an Early Colonization of the New World. In Entering the Americas: Northeast Asia and Beringia Before the Last Glacial Maximum, edited by D. B. Madsen, pp. 187- 238. University of Utah Press, Salt Lake City.

Schurr T.G. and S.T. Sherry 2004 Mitochondrial DNA and Y-chromosome diversity and the peopling of the Americas. American Journal of Human Biology 16(4): 420-39.

Shott, M.J. 1986 Settlement Mobility and Technological Organization: An Ethnographic Examination. Journal of Anthropological Research 42:15-51.

1994 Size and Form in the Analysis of Flake Debris: Review of Recent Approaches. Journal of Archaeological Method and Theory 1:69:-110.

Stafford, T.W., P.E. Hare, L. Currie, A.J. Jull and D.J. Donahue 1991 Accelerator Radiocarbon Dating at the Molecular Level. Journal of Archaeological Science 18:35-72.

Steffian, A.F., E. P. Eufemio and P.G. Saltonstall 2002 Early Sites and Microblade Technologies from the Kodiak Archipelago. Anthropological Papers of the University of Alaska NS 2(1):1-38.

Steward, J. 1972 Theory of Culture Change. The Methodology of Multilinear Evolution. University of Illinois Press, Chicago.

Stuiver, M., G.W. Pearson and T.F. Brasiunas 1986 Radiocarbon Age Calibration of Marine Samples back to 9000 cal yr BP. Radiocarbon 28:980–1021.

Stuiver, M. and P.J. Reimer 1993 Extended 14C Data Base and Revised CALIB 3.0 14C Age Calibration Program. Radiocarbon 35:215-230.

Surovell, T.A. and P.J. Brantingham 2007 A Note on the Use of Temporal Frequency Distributions in Studies of Prehistoric Demography. Journal of Archaeological Science 34:1868-1877.

Surovell, T.A., N.M. Waguespack and P.J. Brantingham 2005 Global Archaeological Evidence for Proboscidean Overkill. Proceedings of the National Academy of Sciences of the United States America 102: 6231-6236.

223

Tedor, R. 2006 Geoarchaeological Investigations in the Susitna Lowlands, South-Central Alaska. unpublished BA thesis, Department of Anthropology, University of Alaska, Anchorage.

Ten Brink, N.W. 1984 North Alaska Range Project Final Report on 1978-1982 Geo-Archeological Studies. Report to the National Geographic Society and the National Park Service. Department of Geology, Grand Valley State College, Allendale, Michigan.

Ten Brink, N.W. and C.F. Waythomas 1985 Late Wisconsin Glacial Chronology of the Northcentral Alaska Range: A Regional Synthesis and its Implications for Early Human Settlements. In North Alaska Range Early Man Project, edited by W. Swanson, pp. 15-32. National Geographic Society Research Reports.

Thorson, R.M. 2006 Artifact Mixing at the Dry Creek Site, Interior Alaska. Anthropological Papers of the University of Alaska 4(1):1-10.

Thorson, R.M. and T.D. Hamilton 1977 Geology of the Dry Creek Site: A Stratified Early Man Site in Interior Alaska. Quaternary Research 7:149-176.

Torrence, R. 1983 Time Budgeting and Hunter-Gatherer Technology. In Hunter-Gatherer Economy in Prehistory, edited by G. Bailey, pp 11-22. Cambridge University Press, Cambridge.

1989 Retooling: Toward a Behavioral Theory of Stone Tools. In Time, Energy and Stone Tools. edited by R. Torrence. pp. 57–66. Cambridge University Press, Cambridge.

Treganza, A.E. 1964 An Archeological Survey of Mount McKinley National Park. University of Alaska and U.S. Department of the Interior, National Park Service, Contract No. 14-10-0434-1547.

Tukey, J.W. 1977 Exploratory Data Analysis. Addison-Wesley, Boston.

Ugan, A. and D. Byers 2007 Geographic and Temporal Trends in Proboscidean and Human Radiocarbon Histories During the Late Pleistocene. Quaternary Science Reviews 26(25- 28):3058-3080. 224

VanderHoek, R. 2005 Cultural Resources Plan for the Denali Highway Lands, Central Alaska. Office of History and Archaeology Report No. 112. State of Alaska, Department of Natural Resources, Division of Parks and Recreation, Anchorage.

VanderHoek, R. and R. Myron 2004 Cultural Remains from a Catastrophic Landscape: An Archaeological Overview and Assessment of Aniakchak National Monument and Preserve. Research Report AR/CRR-2004-47, National Park Service, U.S. Department of the Interior, Anchorage.

VanderHoek, R. R.M. Tedor, C.E. Holmes and D. McMahan 2007a Cultural Materials Recovered from Ice Patches in the Denali Highway Region, Central Alaska, 2003-2005. Alaska Journal of Anthropology 5(2) 175-200.

VanderHoek, R, B.T. Wygal, C.E. Holmes, D. McMahan and R.M. Tedor 2005 Ice Patch Atlatls, Frost Boils and Historic Trails: The OHA 2004 Field Season in the Denali Blocks, Denali Highway Region, Alaska. Paper presented at the 32th Annual Meetings of the Alaskan Anthropological Association, Anchorage.

VanderHoek, R, B.T. Wygal, R. Tedor and C.E. Holmes 2007b Survey and Monitoring of Ice Patches in the Denali Highway Region, Central Alaska, 2003-2005. Alaska Journal of Anthropology 5(2):67-86.

Vasil'ev, S., Ya. Kuzmin, L. Orlova and V. Dementiev 2002 Radiocarbon-Based Chronology of the Paleolithic in Siberia and its Relevance to the Peopling of the New World. Radiocarbon 44 (2).

Veltre, D.W., A. McCartney and M. Veltre and J. Aigner 1984 An Archaeological Site Survey of Amaknak and Unalaska Islands, Alaska. Alaska Division of Parks and Outdoor Recreation. Historic Preservation Fund Grant. Project 84197, Anchorage.

Viereck, L.A. 1966 Plant Succession and Soil Development on Gravel Outwash of the Muldrow Glacier, Alaska. Ecological Monographs 36(3):182-199.

Vinson, D. 1988 Preliminary Report on Faunal Identifications from Trail Creek Caves, In The Bering Land Bridge National Preserve: An Archaeological Survey, Vol. 1, edited by J. Schaaf, pp. 410-438. National Park Service, Anchorage.

1993 Taphonomy and Archaeology of the Trail Creek Caves, Western Alaska. Master’s Thesis, University of Alaska, Fairbanks.

225

Waguespack, N.M. and T.A. Surovell 2003 Clovis Hunting Strategies, or How to Make Out on Plentiful Resources. American Antiquity 68(2):333-352.

Wahrhaftig, C. 1958 Quaternary Geology of the Nenana River Valley and Adjacent Part of the Alaska Range. Geological Survey. Professional Paper No. 293.

Walker, M.C., S. Björck, J.J. Lowe, L. Cwynar, S. Johnsen, K. Knudsen, B. Wohlfarth and the INTIMATE group 1999 Isotopic ‘Events’ in the GRIP Ice Core: A Stratotype for the Late Pleistocene. Quaternary Science Reviews 18:1143-1150.

Walker, D.A., J.G. Bockheim, F.S. Chapin III, W. Eugster, F. E. Nelson, C. L. Ping 2001 Calcium-Rich Tundra, Wildlife, and the 'Mammoth Steppe'. Quaternary Science Reviews 20(1-3):149-163.

Walker, T. 2005 Kantishna: Miners, Mushers and Mountaineers - The Story Behind Mt. McKinley National Park. Pictorial Histories Publishing Company, Inc., Missoula.

Wan, E. 2007 Lab report SU05-TAL092-U10L4. Menlo Park: United States Geological Survey, Tephrochronology Laboratory. Report on file.

Waterbolk, H.T. 1971 Working with Radiocarbon Dates. Proceedings of the Prehistoric Society 37:15- 33.

Waters, M.R. 1992 Principals of Geoarchaeology: A North American Perspective. The University of Arizona Press, Tucson.

Waters, M.R. and T. Stafford 2007 Redefining the Age of Clovis: Implications for the Peopling of the Americas. Science (315):1122-1126.

Weertman, J. 1962 Catastrophic Glacier Advances. International Association of Hydrological Sciences 58:31-39.

Weninger, B. 1986 High-Precision Calibration of Archaeological Radiocarbon Dates Acta Interdisciplinaria Archaeologica IV. Nitra: 11-53.

226

Weninger, B. and O. Jöris 2004 Glacial Radiocarbon Calibration. The CalPal Program. In Radiocarbon and Archaeology, edited by T. Higham, C. B. Ramsey and C. Owen. Fourth International Symposium, Oxford.

2008 A 14C Age Calibration Curve for the last 60 ka: the Greenland-Hulu U/Th Timescale and its Impact on Understanding the Middle to Upper Paleolithic Transition in Western Eurasia. Journal of Human Evolution 55:772-781.

Weninger, B., O. Jöris and U. Danzeglocke 2005 The Cologne Radiocarbon Calibration and Paleoclimate Research Package (CALPAL). University of Cologne.

Werner, A. 1982 Glacial Geology of the McKinley River Area, Alaska: With an Evaluation of Various Relative Age Dating Techniques. Unpublished Master’s thesis, Department of Geology, Southern Illinois University, Edwardsville.

Werner, A., K. Wright and J. Child 1993 Bluff Stratigraphy along the McKinley River: A Record of late Wisconsin Climate Change. Geological Society of America Abstracts with Programs 25:224.

West, F.H. 1965 Excavations at Two Sites on the Teklanika River, Mount McKinley National Park, Alaska. National Park Service and University of Alaska Report, Anchorage.

1967 The Donnelly Ridge Site and the Definition of an Early Core and Blade Complex in Central Alaska. American Antiquity 32(3):360-382.

1972 Archaeological and Paleoecological Research in the Tangle Lakes, Central Alaska, 1966-1972. Manuscript on file at Office of History and Archaeology, Anchorage.

1973 A Report of Archaeological Activity in the Tangle Lakes, Central Alaska in 1973. Manuscript on file at Office of History and Archaeology, Anchorage.

1974 The Significance of Typologically Early Site Collections in the Tangle Lakes, Central Alaska: A Preliminary Consideration. In International Conference on the Prehistory and Paleoecology of Western North American Arctic and Subarctic, edited by S. Raymond and P. Schlederman, pp. 217-238. University of Calgary Archaeological Association, Calgary.

1975 Dating the Denali Complex. Arctic Anthropology 12(1):76-81.

227

1981 The Archaeology of Beringia. Columbia University Press, New York.

1984 Old World Affinities of Archaeological Complexes from Tangle Lakes, Central Alaska. In Beringia in the Cenozoic Era, edited by V.L. Kontrimavichus, pp. 571- 596. Oxonian Press, New Delhi.

1996a The Archaeological Evidence. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 537-560. University of Chicago Press, Chicago.

1996b Teklanika West. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 332-343. University of Chicago Press, Chicago.

1996c Amphitheater Mountain Sites. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 405-408. University of Chicago Press, Chicago.

1996d American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West. University of Chicago Press, Chicago.

1996e Donnelly Ridge. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 302-307. University of Chicago Press, Chicago.

West, F.H., B.S. Robinson and M.L. Curran 1996a Phipps Site. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 381-386. University of Chicago Press, Chicago.

West, F.H., B.S. Robinson and R.G. Dixon 1996b Sparks Point. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 394-398. University of Chicago Press, Chicago.

West, F.H., B.S. Robinson and C. West 1996c Whitmore Ridge. In American Beginnings: The Prehistory and Paleoecology of Beringia, edited by F.H. West, pp. 386-394. University of Chicago Press, Chicago.

Westgate, J.A., Stemper, B.A. and T.L. Péwé 1990 A 3 M.Y. Record of Pliocene–Pleistocene Loess in Interior Alaska. Geology 18, 858–861.

228

White, L. 1959 The Evolution of Culture. McGraw Hill, New York.

White, T.E. 1954 Observations on the Butchering Technique of Some Aboriginal Peoples. American Antiquity (19)3:254-264.

Whittington, S.L. and B. Dyke 1984 Simulating Overkill: Experiments with the Mosimann and Martin Model. In Quaternary Extinctions: A Prehistoric Revolution. edited by P. S. Martin, R. G. Klein, pp. 451-465. University of Arizona Press, Tucson.

Wilson, M.C. and J.A. Burns 1999 Searching for the Earliest Canadians: Wide Corridors, Narrow Doorways, Small Windows. In Ice Age Peoples of North America: Environments, Origins and Adaptations of the First Americans, edited by R. Bonnichsen and K.L. Turnmire, pp. 213-248. Oregon State University Press, Corvallis.

Winterhalder, B., W. Baillargeon, F. Cappelletto, I. R. J. Daniel Jr. and C. Prescott 1988 The Population Ecology of Hunter-Gatherers and their Prey. Journal of Anthropological Archaeology 7:289-328.

Workman, W. 1978 Prehistory of the Aishinik-Kluane Area, Southwest Yukon Territory. National Museum of Man, Ottawa.

1998 Coast to Coast: Prehistoric Maritime Cultures in the North Pacific. Arctic Anthropology 35(1):361-370.

Wygal, B.T. 2003 The Earliest Alaskans and the Earliest Americans Theme Study. Unpublished Master’s thesis, Department of Anthropology, University of Alaska, Anchorage.

2007a Microblade/Non-Microblade Dichotomy in Southcentral Alaska. Paper presented at the 72rd Annual Meetings of the Society for American Archaeology, Austin.

2007b Microblade/Non-Microblade Dichotomy in Southcentral Alaska. Paper presented at the 34th Annual Meetings of the Alaskan Anthropological Association, Fairbanks.

2007c Cultural Resource Exploration in Denali National Park and Preserve: 2007 Annual Report. National Park Service, Denali Park, Alaska.

229

2008a Following the Ice? Early Holocene Adaptations in Southcentral Alaska. Paper presented at the 73rd Annual Meetings of the Society for American Archaeology, Vancouver.

2008b The Risk of Global Climate Change to Cultural Resources in Denali National Park and Preserve: A Report on Current Efforts and Future Needs in Archeological Monitoring and Inventory. National Park Service, Denali Park, Alaska.

in press The Microblade/Non-Microblade Dichotomy: Climatic Implications, Toolkit Variability and the Role of Tiny Tools in Eastern Beringia. In From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia, edited by T. Goebel and I. Buvit. Texas A&M University Press, College Station.

in review Analogous Scenarios: Deglaciation and the Human Colonization of Northern Europe. Alaska Journal of Anthropology.

Wygal, B.T. and T. Goebel 2006 The Early Holocene Colonization of Southcentral Alaska. Paper presented at the 33rd annual Alaska Anthropological Association, Kodiak.

Wygal, B.T. and K.E. Krasinski 2006 The Archaeology of Southcentral Alaska and the Microblade Dichotomy. Paper presented at the Workshop on Pleistocene Human Colonization of Arctic and Subarctic Siberia and Beringia, Center for the Study of the First Americans, Texas A&M University, International Quaternary Association and National Science Foundation.

Wygal, B.T. and F. Seager-Boss 2005 The Holocene Peopling of Southcentral Alaska: New Evidence from Trapper Creek. Paper presented at the 32th Annual Meetings of the Alaskan Anthropological Association, Anchorage.

Yesner, D.R. 1996 Human Adaptation at the Pleistocene-Holocene Boundary (circa 13,000 to 8,000 B.P.) in Eastern Beringia. In Humans at the End of the Ice Age: Archaeology of the Pleistocene-Holocene Transition, edited by L.G. Straus, Berit Eriksen, J.M. Erlansdson and D.R. Yesner, pp. 243-276. Plenum Press, New York.

1998 Origins and Development of Maritime Adaptations in the Northwest Pacific Region of North America: A Zooarchaeological Perspective. Arctic Anthropology 35(1):204-222.

230

2000 Human Colonization of Eastern Beringia and the Question of Mammoth Hunting. Proceedings of the International Conference on Mammoth Site Studies, Occasional Papers in Anthropology No 22:69-84. University of Kansas, Lawrence.

2001 Human Dispersal into Interior Alaska: Antecedent Conditions, Mode of Colonization and Adaptations. Quaternary Science Reviews 20:315-327.

Yesner, D.R., C.M. Barton, G.A. Clark and G.A. Pearson 2004 Peopling of the Americans and Continental Colonization: A Millennial Perspective. In The Settlement of the American Continents: A Multidisciplinary Approach to Human Biogeography, edited by C.M. Barton, G.A. Clark, D.R. Yesner and G.A. Pearson, pp.196-214. University of Arizona Press, Tucson.

Yesner, D.R., C.E. Holmes and K.J. Crossen 1992 Archaeology and Paleoecology of the Broken Mammoth Site, Central Tanana Valley, Interior Alaska. Current Research in the Pleistocene 9:53-57.

Yesner, D.R. and G. Pearson 2002 Microblades and Migrations: Ethnic and Economic Models for the Peopling of the Americas. In Thinking Small: Global Perspectives on Microlithization, edited by R.G. Elston and S.L. Kuhn, pp. 133-161. Anthropological Papers of the American Anthropological Association. No. 12.

Yesner, D., G. Pearson and D.E. Stone 2000 Additional Organic Artifacts from the Broken Mammoth Site, Big Delta, Alaska. Quaternary Research in the Pleistocene 17:87-90.

Yi, S. and G. Clark 1985 The ‘Dyuktai Culture’ and New World Origins. Current Anthropology 26(1):1- 20.

Zinck, B. and T. Zinck 1976 Survey of Archaeological Sites, Tangle Lakes Archaeological District near Paxson, Alaska, edited by J. Beck, BLM Anchorage. Western Interstate Commission for Higher Education, Boulder.

231

Curriculum Vitae Brian T. Wygal Archeologist

EDUCATION PhD Anthropology (May 2009) University of Nevada, Reno. Dissertation Title: The Prehistoric Colonization of Southcentral Alaska: Human Adaptations in a Post Glacial World

MA Anthropology (2003) University of Alaska, Anchorage. Thesis topic: The Earliest Alaskans

BS Anthropology (1998) Kansas State University. Emphasis: cultural anthropology and prehistoric Archeology of the Great Plains

RESEARCH INTERESTS Arctic, Subarctic and Great Plains Archeology, hunter-gatherer adaptations, evolutionary ecology, prehistoric technology and climate change.

ACADEMIC APPOINTMENTS Instructor (2006-2008) University of Nevada, Reno: Introduction to Archeology.

Instructor (2005-2006) University of Nevada, Reno: Field School in Prehistoric Archeology, Trapper Creek, Alaska.

Instructor (2005-2006) University of Nevada, Reno: Introduction to Physical Anthropology.

Adjunct Faculty (2004) University of Alaska, Palmer: Rise of Civilization.

Research Associate (2003) University of Alaska, Anchorage: Western Arctic National Park Lands.

Teaching Associate (2002-2003) University of Alaska, Anchorage: Introduction to Anthropology and Rise of Civilization.

Teaching Assistant (2001-2002) University of Alaska, Anchorage: Introduction to Anthropology and Rise of Civilization.

Instructor (2000-2001) Kansas State University: Introduction to Cultural Anthropology Recitation.

Teaching Assistant (1999-2000) Kansas State University: Field and Lab Techniques in Archeology, Medicine Creek, Nebraska.

232

EMPLOYMENT Principal Investigator (2006-2009) Archeologist, United States Department of the Interior, National Park Service, Denali National Park and Preserve: Comprehensive Archeological Survey of Denali National Park and Preserve.

Research Supervisor (2005) Matanuska Susitna Borough, Cultural Resources Division, (PI Seager-Boss).

Field Supervisor (2004) Alaska State Department of Natural Resources, Office of History and Archeology, (PI VanderHoek).

Field Supervisor (2004) Matanuska-Susitna Borough, Cultural Resources Division, (PI Seager-Boss).

Archeology Intern (2004) National Park Service, Alaska Regional Office: Archeological Sites Management Information System for Denali National Park and Preserve, (supervisor Bender).

Field Technician (2003) Museum of the Aleutians Excavation of the Amaknak Bridge site, Dutch Harbor, Alaska, (PIs Knecht and Davis).

Laboratory Technician (2002-2003) Department of the Interior, National Park Service, Western Arctic National Parklands, (PI Gal, R).

Archeology Intern (2000-2003) Department of the Interior, National Park Service, Alaska Regional Office: The Earliest Americans Theme Study in Alaska, (supervisor Saleeby).

Field Supervisor (2002) Colorado State University-United States Army, Fort Greeley, Alaska, (PI Hedman).

Field Supervisor and Laboratory Technician (2000-2001) Kansas State University, Upper Republican River Basin, Medicine Creek, Nebraska (PI Roper).

Field Technician (5/2000, 5/2001) University of Kansas: Kirwin Wildlife Refuge , Kansas (PI Logan).

Laboratory Assistant (2000) Kansas State University: NAGRPA Compliance and Repatriation Project, (supervisor Roper).

Field Technician (1999) Kansas State University: Upper Republican River Basin, Medicine Creek, Nebraska (PI Roper).

Laboratory Technician (1998-1999) Kansas State Historical Society, Archeology Laboratory, (supervisors Stein and Wulfkuhle). 233

PUBLICATIONS Wygal, BT (in review) Analogous Scenarios: Deglaciation and the Human Colonization of Northern Europe. Alaska Journal of Anthropology.

Wygal, BT and Goebel, T (in prep) Early Holocene Settlement of Southcentral Alaska: New Results from the Trapper Creek Archaeological Sites. intended for Arctic.

Wygal, BT (in press) The Microblade/Non-Microblade Dichotomy: Climatic Implications, Toolkit Variability and the Role of Tiny Tools in Eastern Beringia. In From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia, edited by T. Goebel, T and I. Buvit. Texas A&M Press, College Station.

VanderHoek, R, Wygal, BT, Tedor, R and Holmes, C (2007) Survey and Monitoring of Ice Patches in the Denali Highway Region, Central Alaska, 2003- 2005. Alaska Journal of Anthropology 5(2):67-86.

Saleeby, B and Wygal, BT (2005) Old is Getting Older. Alaska Park Science. National Park Service, Anchorage Support Office, Anchorage.

Roper, D and Wygal, BT (2002) The Spatial Component of the Western Clovis Chronology. Current Research in the Pleistocene 19:76-78.

GOVERNMENT REPORTS Wygal, BT (2009) Prehistoric Upland Lithic Procurement and Hunting Strategies in Denali National Park and Preserve, Central Alaska. National Park Service, Denali Park, Alaska.

Wygal, BT (2009) Archeological Survey Proposal for Denali National Park and Preserve, Summer 2009 Field Plan. National Park Service, Denali Park, Alaska.

Wygal, BT (2008) Cultural Resource Exploration in Denali National Park and Preserve: 2008 Annual Report. National Park Service, Denali Park.

Wygal, BT (2008) The Risk of Global Climate Change to Cultural Resources in Denali National Park and Preserve: A Report on Current Efforts and Future Needs in Archeological Monitoring and Inventory. National Park Service, Denali Park, Alaska.

Wygal, BT (2008) Archeological Survey Proposal for Denali National Park and Preserve, Summer 2008 Field Plan. National Park Service, Denali Park, Alaska.

234

Saleeby, B and Wygal, BT (2007) The Earliest American Theme Study for Alaska. National Register of Historic Places Multiple Property Documentation Form, National Park Service, Alaska Regional Office, Anchorage.

Wygal, BT (2007) Cultural Resource Exploration in Denali National Park and Preserve: 2007 Annual Report. National Park Service, Denali Park, Alaska.

Wygal, BT (2007) Archeological Survey Proposal for Denali National Park and Preserve, Summer 2007 Field Plan. National Park Service, Denali Park., Alaska.

Wygal, BT (2006) Cultural Resource Exploration in Denali National Park and Preserve: 2006 Annual Report. National Park Service, Denali Park, Alaska.

Goebel, T and Wygal, BT (2006) The Early Holocene Colonization of Southern Alaska, National Science Foundation, Office of Polar Programs, Final Report for Grant #052055.

GRANT SUPPORT Goebel, T (PI) and Wygal, BT (co-PI) National Science Foundation, Office of Polar Programs, Arctic Social Sciences, #052055 (2005) $74,808.

Goebel, T (PI) and Wygal, BT (co-PI) University of Nevada, Reno, Academy for the Environment Research Grant, Early Holocene Colonization of Southcentral Alaska (2006) $5000.

Wygal, BT (PI) and Goebel, T (co-PI) Alaska Humanities Forum Research Grant, The Prehistoric Colonization of Southcentral Alaska (2005) $4000.

Wygal, BT (PI) University of Nevada, Reno, Graduate Student Association Research Grant, The Early Holocene Colonization of Southcentral Alaska (2005) $2500.

CONFERENCES AND PUBLISHED ABSTRACTS Wygal, BT (2009) Analogous Scenarios: Deglaciation and the Human Colonization of Northern Europe. Paper presented at the 36th Annual Meetings of the Alaskan Anthropological Association, Juneau.

Wygal, BT (2009) Archeological Survey of Denali National Park and Preserve: Assessments, Models, and Hypotheses. Paper presented at the 36th Annual Meetings of the Alaskan Anthropological Association, Juneau.

Wygal, BT (2008) Following the Ice? Early Holocene Adaptations in Southcentral Alaska. Paper presented at the 73rd Annual Meetings of the Society for American Archeology, Vancouver (invited lecture).

235

Wygal, BT (2008) Results of the 2007 Archaeological Survey of Denali National Park & Preserve. Paper presented at the 35th Annual Meetings of the Alaskan Anthropological Association, Anchorage.

Coffman, S and Wygal, BT, (2008) Early Holocene Lithic Reduction from the Bull River and Costello Creek Area of Denali National Park. Paper presented at the 35th Annual Meetings of the Alaskan Anthropological Association, Anchorage. Wygal, BT (2007) Microblade/Non-Microblade Dichotomy in Southcentral Alaska. Paper presented at the 72rd Annual Meetings of the Society for American Archeology, Austin.

Wygal, BT (2007) Microblade/Non-Microblade Dichotomy in Southcentral Alaska. Paper presented at the 34th Annual Meetings of the Alaskan Anthropological Association, Fairbanks.

Wygal, BT and Krasinski, K (2006) The Archeology of Southcentral Alaska. Workshop on Pleistocene Human Colonization of Arctic and Subarctic Siberia and Beringia. Paper presented at the Center for the Study of the First Americans, Texas A&M University, International Quaternary Association and National Science Foundation sponsors (invited lecture).

Wygal, BT and Goebel, T (2006) The Early Holocene Colonization of Southcentral Alaska. Paper presented at the 33th Annual Meetings of the Alaskan Anthropological Association, Kodiak.

Wygal, BT and Seager-Boss, F (2005) The Holocene Peopling of Southcentral Alaska: New Evidence from Trapper Creek. Paper presented at the 32th Annual Meetings of the Alaskan Anthropological Association, Anchorage.

VanderHoek, R, Wygal, BT, Holmes, C, McMahan, D, and Tedor, R (2005) Ice Patch Atlatls, Frost Boils, and Historic Trails: The OHA 2004 Field Season in the Denali Blocks, Denali Highway Region, Alaska. Paper presented at the 32th Annual Meetings of the Alaskan Anthropological Association, Anchorage.

Wygal, BT and Seager-Boss, F (2005) Early Holocene Colonization of Southcentral Alaska. Paper presented at the 13th Annual Arctic Conference, University of California-Davis.

Wygal, BT and Gal, R (2004) The Last Day Site (XMH-497): A Notched Projectile Point and Microblade Technology in the Early Holocene of Northwest Alaska. Paper presented at the 31th Annual Meetings of the Alaskan Anthropological Association, Whitehorse.

236

Wygal, BT (2003) The Earliest Alaskans and the Earliest Americans Theme Study. Paper presented at the 30th Annual Meetings of the Alaskan Anthropological Association, Fairbanks.

PROFESSIONAL DEVELOPMENT Wygal, BT and Reuther, J (2009) Co-organizer and chairperson of Humans and their Environment: Technology and Subsistence in Arctic and Subarctic Landscapes. Alaska Journal of Anthropology. Journal Volume and Symposium presented at the 36th Annual Meetings of the Alaska Anthropological Association, Juneau.

Wygal, BT (2008) Organizer and chairperson of The Archeology of Alpine and Montane Environments. Symposium presented at the 35th Annual Meetings of the Alaska Anthropological Association, Anchorage.

Wygal, BT and Yesner, D (2002) Co-organizer of American Quaternary Association, Biennial Conference and Interior Alaska Paleoindian Site Tour, Anchorage.

COMMUNITY OUTREACH Wygal, BT (2008) The Archeology of Denali and Interior Alaska. Denali National Park, Brown Bag Lecture Series, Denali Park, Alaska.

Wygal, BT (2007) Prehistoric Archeology of Denali: Prehistoric Landuse in Interior Alaska and the 2007 Survey Results. Denali National Park, Brown Bag Lecture Series, Denali Park, Alaska.

Wygal, BT (2007) The Holocene Peopling of Southcentral Alaska. Public Presentation, National Park Service Ranger Station, Talkeetna, Alaska.

Wygal, BT, Seager-Boss, F, Krasinski, K, and Louderback, L (2006) Archeology Outreach, Presented to the Trapper Creek Elementary School, Trapper Creek, Alaska.

Wygal, BT, (2006) Prehistoric Archeology of Alaska. Presented to the Hill City Rotary Club, Hill City, Kansas.

Wygal, BT, Seager-Boss, F, Krasinski, K, and Stone, D (2005) Public participation in the excavation of an early prehistoric site, Trapper Creek, Alaska.

Wygal, BT, (2005) Prehistoric Archeology of Alaska. Presented to the Amateur Archeology Association of Nevada at the Desert Research Institute, Reno.

Wygal, BT, (2004) Prehistoric Archeology of Alaska, Anthropology Club Presentation Series, University of Nevada, Reno.

237

Wygal, BT, Krasinski, K, and Blechl, C (2003) Amaknak Bridge Excavations. Anthropology Club Presentation Series, University of Alaska, Anchorage.

ADDITIONAL EXPERIENCE Research Volunteer (1999) Kansas State Historical Society, Penokee Stone Effigy and Sumner Battle surveys (advisor Rowlison).

Field School Student (1998) Kansas State University, Upper Republican, semi- sedentary hunter-horticultural societies, survey and site excavation in Medicine Creek, Nebraska (PI Roper).

PROFESSIONAL AFFILIATIONS Society of American Archeology Alaska Anthropological Association

FOREIGN LANGUAGE Russian Language Immersion Program (May 2008), Center of Russian Language and Culture, Saint Petersburg State University, Russia.

AWARDS UNR: The Honor Society of Phi Kappa Phi (2009).

UNR: Graduate Student Association, Presentation Competition (2008) 1st place.

UNR: Graduate Student Association, Graduate Travel Awards (2007-2009).

UNR: Graduate Student Association, Presentation Competition (2006) 2nd place.

UAA: Anthropology Graduate Travel Award (2002) Nebraska Stone Effigy Survey.