Cultural Resources Data Recovery at Bayou Park (8OK898), Eglin Air Force Base, Okaloosa County, Florida
Descrição do Produto
FINAL REPORT
CULTURAL RESOURCES DATA RECOVERY AT BAYOU PARK (8OK898), EGLIN AIR FORCE BASE, OKALOOSA COUNTY, FLORIDA
CONTRACT NO. EAFB‐62221‐008 TASK ORDER NO. CR‐11‐0051 PREPARED FOR MID‐BAY BRIDGE AUTHORITY AND 96TH CIVIL ENGINEERING GROUP, EGLIN AIR FORCE BASE PREPARED BY ROBERT J. AUSTIN, CHRISTOPHER MICKWEE, AND JOSHUA M. TORRES WITH CONTRIBUTIONS BY ANNA R. DIXON, IRVY R. QUITMYER, DENNIS K. WARDLAW, BRIAN WORTHINGTON, AND CHAD YOST
________________________________ ROBERT J. AUSTIN, PHD, RPA PRINCIPAL INVESTIGATOR
SOUTHEASTERN ARCHAEOLOGICAL RESEARCH, INC. WWW.SEARCHINC.COM
SEARCH PROJECT NO. 2614‐11010V AUGUST 2014
This page intentionally left blank
TABLE OF CONTENTS List of Figures ........................................................................................................................... vii List of Tables .......................................................................................................................... xvii List of Acronyms ..................................................................................................................... xxi Acknowledgments ................................................................................................................ xxiii SEARCH Project Participants ................................................................................................ xxiv Executive Summary ............................................................................................................... xxv Chapter 1: Introduction ..................................................................................................... 1 Site Location and Present Condition ................................................................................... 2 Summary of Previous Investigations................................................................................... 2 Summary of Phase III Results ............................................................................................ 24 Conclusion ......................................................................................................................... 30 Chapter 2: Environmental and Archaeological Contexts ....................................... 31 Regional Environmental Setting ....................................................................................... 31 Paleoenvironmental Conditions ....................................................................................... 44 Prehistoric Overview ......................................................................................................... 49 Conclusion ......................................................................................................................... 68 Chapter 3: Research Design and Methods ................................................................. 69 Research Design ................................................................................................................ 69 Field Methods ................................................................................................................... 77 Laboratory Methods ......................................................................................................... 95 NAGPRA‐Related Human Remains ................................................................................... 95 Curation ............................................................................................................................ 96 Appendices ........................................................................................................................ 97 Chapter 4: Stratigraphy and Chronology .................................................................. 109 Description of Strata ....................................................................................................... 109 Site Chronology ............................................................................................................... 123 Temporally Diagnostic Artifacts ...................................................................................... 127 Conclusion ....................................................................................................................... 134 Appendices ...................................................................................................................... 139 Chapter 5: Ground‐Penetrating Radar Survey by Dennis K. Wardlaw ................... 147 Methods .......................................................................................................................... 147 GPR Survey Results ......................................................................................................... 149 Discussion of Results ....................................................................................................... 160 Conclusion ....................................................................................................................... 163 Chapter 6: Midden Deposits and Features .............................................................. 165 Midden Deposits ............................................................................................................. 167 Features .......................................................................................................................... 171 Lipid Residue Analysis of Selected Features ................................................................... 245 iii
Functional Interpretations .............................................................................................. 249 Spatial and Temporal Distributions ................................................................................ 255 Structures ........................................................................................................................ 261 Discussion........................................................................................................................ 262 Conclusion ....................................................................................................................... 274 Appendices ...................................................................................................................... 275 Chapter 7: Baked‐Clay Objects .................................................................................... 291 Background ..................................................................................................................... 291 Bayou Park Baked‐Clay Object Typology ........................................................................ 294 Analysis Methods ............................................................................................................ 298 Assemblage Description .................................................................................................. 300 X‐Ray Fluorescence Spectrometry Analysis .................................................................... 311 Spatial Distribution ......................................................................................................... 314 Functional Interpretations .............................................................................................. 316 Technological Interpretations ......................................................................................... 322 Temporal Comparisons ................................................................................................... 323 Conclusions ..................................................................................................................... 328 Appendix ......................................................................................................................... 331
Chapter 8: Phytolith and Starch Analysis of a Possible Griddle Fragment from Bayou Park by Chad Yost............................................................... 353
Background ..................................................................................................................... 353 Methods .......................................................................................................................... 353 Results ............................................................................................................................. 354 Summary and Conclusions .............................................................................................. 357 Chapter 9: Prehistoric Pottery ..................................................................................... 359 Methods .......................................................................................................................... 359 Results ............................................................................................................................. 360 Conclusion ....................................................................................................................... 367 Appendix ......................................................................................................................... 369 Chapter 10: Lithic Artifacts: Resource Exploitation and Use .............................. 375 Summary of Previous Research ...................................................................................... 375 Analysis Methods ............................................................................................................ 378 Lithic Raw Materials ........................................................................................................ 385 Artifact Descriptions ....................................................................................................... 390 Artifact Function ............................................................................................................. 432 Temporal and Spatial Variation ...................................................................................... 434 Technological Organization ............................................................................................ 444 Assemblage Diversity ...................................................................................................... 448 Discussion........................................................................................................................ 448 Conclusion ....................................................................................................................... 457 Appendices ...................................................................................................................... 459 Chapter 11: Bone, Antler, and Shell Artifacts ......................................................... 493 Analysis Methods ............................................................................................................ 493 iv
Results ............................................................................................................................. 493 Functional Interpretations .............................................................................................. 509 Discussion........................................................................................................................ 510 Conclusions ..................................................................................................................... 511 Appendix ......................................................................................................................... 512 Chapter 12: Faunal Analysis by Brian Worthington ..................................................... 517 Materials and Methods ................................................................................................... 517 Results ............................................................................................................................. 522 Habitats and Exploitation Methods ................................................................................ 552 Seasonality ...................................................................................................................... 561 Comparisons with Other Sites ........................................................................................ 561 Conclusions ..................................................................................................................... 564 Appendices ...................................................................................................................... 566
Chapter 13: The Sclerochronology of Hard Clams (Mercenaria spp.) Excavated from Bayou Park by Irvy R. Quitmyer ................................................... 605
Materials and Methods ................................................................................................... 609 Results ............................................................................................................................. 611 Discussion........................................................................................................................ 614 Conclusions ..................................................................................................................... 615 Appendix ......................................................................................................................... 616
Chapter 14: Macrobotanical Materials from Features at Bayou Park by Anna R. Dixon ............................................................................................................. 619
Research Methods and Objectives ................................................................................. 619 Results ............................................................................................................................. 620 Conclusions ..................................................................................................................... 625
Chapter 15: Interpretation of Archaeological Data: Bayou Park, Elliott’s Point, and Beyond ..................................................................................... 629 Bayou Park: A Site History .............................................................................................. 629
The Bigger Picture: Evaluating Elliott’s Point .................................................................. 647 Future Directions ............................................................................................................ 677 Conclusions ..................................................................................................................... 679 Chapter 16: Conclusion .................................................................................................. 681 References Cited ................................................................................................................... 683 Appendix A: Memorandum of Agreement Appendix B: Updated Florida Master Site File Form Appendix C: FDHR Survey Log Sheet
v
This page intentionally left blank.
vi
LIST OF FIGURES Figure 1.1. Figure 1.2. Figure 1.3. Figure 1.4. Figure 1.5. Figure 1.6. Figure 1.7. Figure 1.8. Figure 1.9. Figure 1.10. Figure 1.11. Figure 1.12. Figure 1.13. Figure 1.14. Figure 1.15. Figure 1.16. Figure 1.17. Figure 1.18. Figure 2.1. Figure 2.2. Figure 2.3. Figure 2.4. Figure 2.5.
Location of 8OK898 within Eglin Air Force Base, Okaloosa County, Florida .... 3 Aerial photograph showing 8OK898 site boundaries within the former Eglin Federal Prison Camp ......................................................................................... 4 Map of 1993 shovel testing at 8OK898 (from Campbell and Meyer 1993)...... 5 Map of 1994 shovel testing at 8OK898 (from Mathews et al. 1994) ............... 7 Map of 1996 shovel testing and test pit excavation at 8OK898 (from Campbell et al. 1996) ........................................................................................ 7 Map showing the location of the 1997 shovel testing and test pit excavation at 8OK898 (from Campbell and Mathews 1997) .............................................. 8 Map showing the location of the 1998 shovel testing at 8OK898 (from Campbell et al. 1998) ........................................................................................ 8 Map showing the location of the 1998 block excavation at 8OK898 (from Campbell et al. 1998) ........................................................................................ 9 Map showing the location of the 2002 shovel testing and test pit excavation at 8OK898 (from Meyer et al. 2002) ............................................ 11 Map showing the location of the 2007 shovel testing and test pit ..................... excavation at 8OK898 (from Morehead et al. 2008) ...................................... 13 Map showing the location of the 2008 test pit excavation northeast of Building 635 at 8OK898 (from Morehead and Aubuchon 2008) .................... 15 Map showing the location of the 2008 test pit excavation southwest of Building 635 at 8OK898 (from Morehead and Aubuchon 2008) .................... 15 Map showing the location of the 2009 shovel testing and test pit excavation at 8OK898 (from Campbell, Morehead, et al. 2009) .................... 17 Map showing the location of the 2010 shovel testing and test pit excavation at 8OK898 (from Morehead et al. 2011) ...................................... 19 Map showing the location of the 2010 monitoring and the area at Building 591 where midden deposits were identified (from Speal and Nolan 2011) .. 20 Map of shovel tests excavated by PTA, 1993‐2010 ........................................ 21 Map of test pits excavated by PTA, 1993‐2010 .............................................. 22 Map showing Operations A, B, and C within site 8OK898 as well as the locations of Strip Areas and Test Units. .......................................................... 25 Portion of USGS Destin, FL 7.5’ quadrangle map (top) showing the location of 8OK898 within Eglin AFB and a topographic map (bottom) showing the elevated hill system on which much of the Bayou Park site rests. ................ 32 Physiographic divisions and subdivisions around 8OK898 and Choctawhatchee Bay (Brooks 1981) ............................................................... 33 Map of geological formations in the southeastern U.S. that contain knappable toolstone ....................................................................................... 36 Portion of 1935 USGS Villa Tasso 15’ quadrangle map showing Postl Lake ... 39 USDA aerial photograph showing Postl Lake in 1955 ..................................... 40 vii
Figure 2.6.
Map of Eglin AFB showing the distribution of wetlands in relation to Bayou Park (8OK898) ................................................................................................. 43 Figure 2.7. Seagrass beds in Choctawhatchee bay ca. 1972 ............................................ 44 Figure 2.8. Reconstructed Holocene sea level fluctuations for the Eglin Air Force Base/Choctawhatchee Bay region .................................................................. 48 Figure 2.9. Location of the Bayou Park site in relation to Poverty Point and other related sites ..................................................................................................... 52 Figure 2.10. Distribution of Late Archaic sites in the Choctawhatchee Bay region ........... 57 Figure 3.1. Model of occupational intensity (after Lightfoot 1984) ................................. 75 Figure 3.2. GPR grids within Operations A, B, and C ........................................................ 78 Figure 3.3. GSSI GPR unit used for data collection at 8OK898 ......................................... 81 Figure 3.4. View to the northeast of Operation A showing locations of the concrete sewer marker, PTA’s 1998 block excavation, and SEARCH’s grid datum ....... 82 Figure 3.5. Relative locations of SEARCH and PTA 1000n, 1000E datum points within Operation A ..................................................................................................... 83 Figure 3.6. View to the “north” of Strip Area K with hydraulic excavator ....................... 86 Figure 3.7. Strip areas and test pit locations in Operation A ............................................ 87 Figure 3.8. Strip areas and test pits locations in Operations A and C .............................. 88 Figure 3.9. Strip areas and test pit locations in Operation B ............................................ 89 Figure 3.10. Excavating TPs 59 and 61 inside demolished Building 588, Operation A ....... 93 Figure 3.11. Block excavation in Strip Area Ja .................................................................... 93 Figure 3.12. Documenting a shell‐filled pit feature (Feature 36) in Strip Area B ............... 94 Figure 3.13. Water screening bulk feature sample through 1.6‐cm (1/16‐inch) mesh screen .............................................................................................................. 96 Figure 4.1. Generalized north‐south profile through Operation A based on profiles in Strip Areas E, D, and K ................................................................................... 111 Figure 4.2. Generalized east‐west profile through Operations A, C, and B based on profiles in Strip Areas .................................................................................... 112 Figure 4.3. East wall profile, Test Pit 59, Operation A .................................................... 113 Figure 4.4. East wall profile, Test Pit 60, Operation A .................................................... 113 Figure 4.5. West wall profile, Test Pit 62, Operation A .................................................. 114 Figure 4.6. West wall profile, Test Pit 63, Operation A .................................................. 114 Figure 4.7. South wall profile, Test Pit 85, Operation A ................................................. 115 Figure 4.8. South wall profile, Test Pit 88, Operation B .................................................. 116 Figure 4.9. South wall profile, Test Pit 94, Operation C .................................................. 117 Figure 4.10. Plot of calibrated radiocarbon ages from Bayou Park .................................. 125 Figure 4.11. Plot of calibrated calendrical dates from Bayou Park .................................. 126 Figure 4.12. Examples of ceramic sherds from Bayou Park .............................................. 128 Figure 4.13. Destin points and “Destin‐like” point from Bayou Park ............................... 131 Figure 4.14. Examples of projectile points manufactured from broken proximal and distal fragments ............................................................................................ 132 Figure 4.15. Examples of diagnostic projectile points from Bayou Park .......................... 133 viii
Figure 4.16. Stratigraphic relationships between temporally diagnostic artifacts and radiocarbon dates ......................................................................................... 138 Figure 5.1. GPR grids within Operations A, B, and C ...................................................... 148 Figure 5.2. Top) GPR survey Grid 1 with rectified image showing 1998 PTA block excavation and mechanical stripping. Bottom) The reflection profile along Transect 201 ........................................................................................ 151 Figure 5.3. Top) Strip Area Ja showing the location of Features 69/104 and 120. Bottom) Amplitude maps from the western portion of Grid 1 .................... 152 Figure 5.4. Reflection profiles from GPR Grid 1 .............................................................. 153 Figure 5.5. Top) Feature 60 located next to a sewer pipe and trench in Strip Area Jb. Bottom) Amplitude map of GPR Grid 1 ........................................................ 154 Figure 5.6. Top) Strip Area E showing the top of the midden. Bottom) Amplitude map of the northern portion of GPR Grid 2 .................................................. 155 Figure 5.7. Reflection profile from GPR Grid 6 ............................................................... 156 Figure 5.8. Top) Strip Area K during mechanical excavation. Bottom) Amplitude plan view of the southern portion of GPR Grid 6 ................................................. 157 Figure 5.9. Top) Profile of Feature 73 in Strip Area K. Bottom) Reflection profile from GPR Grid 6 ............................................................................................ 158 Figure 5.10. Top) Amplitude map from GPR Grid 42. Bottom) Similar disturbances are seen in the profile of Test Pit 86, located just to the southeast of GPR Grid 42 ........................................................................................................... 159 Figure 5.11. Top) A reflection profile from GPR Grid 14 with an area of disturbance just below the ground surface highlighted. Bottom) The disturbance was confirmed through mechanical excavation of Strip Area X ................... 161 Figure 5.12. Side‐by‐side comparisons of plan views and profiles of Features 119 and 120 ......................................................................................................... 162 Figure 6.1. View to the northeast of Operation A at Bayou Park showing Strip Areas K and L (upper left), Ja (lower left), and E (lower right) in various stages of work. Strip Areas F and D (center right) have been backfilled and graded ........................................................................................................... 166 Figure 6.2. View to the east of Strip Area Ja showing large pit features exposed during stripping ............................................................................................. 166 Figure 6.3. View to the east of Strip Area Ja showing large pit features (left) and the outlines of PTA’s 1998 block excavation (right) ........................................... 167 Figure 6.4. Estimated extent of midden based on PTA’s 1998 data recovery project (from Campbell et al. 1998) .......................................................................... 168 Figure 6.5. East wall profile of TP60 showing discontiguous nature of shell in “Stratum” IVb ................................................................................................ 169 Figure 6.6. East wall profile of TP58 showing Stratum IVc ............................................. 169 Figure 6.7. View to the southeast of Strip Area B showing dark sediments associated with the top of Stratum IVa. Arrows point to isolated shell deposits (Stratum IVb) ................................................................................................. 170 ix
Figure 6.8. Figure 6.9. Figure 6.10. Figure 6.11. Figure 6.12. Figure 6.13. Figure 6.14. Figure 6.15. Figure 6.16. Figure 6.17. Figure 6.18. Figure 6.19. Figure 6.20. Figure 6.21. Figure 6.22. Figure 6.23. Figure 6.24. Figure 6.25. Figure 6.26. Figure 6.27. Figure 6.28. Figure 6.29. Figure 6.30. Figure 6.31. Figure 6.32. Figure 6.33. Figure 6.34.
Profile of north wall of Strip Area D showing dense midden deposits and a large pit feature at far right ....................................................................... 170 Mapping of Stratum IVc in Strip Area A. View is to the northeast .............. 171 Strip areas and midden deposits (identified and interpreted) in Operation A ................................................................................................... 172 Topographic map of Operation A with midden boundaries ......................... 173 Plan and cross‐section shapes of features.................................................... 176 General distribution of features in Operations A and C ............................... 177 General distribution of features in Operation B ........................................... 178 Features identified in Strip Area A, Operation A .......................................... 179 Features identified in Strip Area B, 3.40‐3.27 m amsl, Operation A ............ 181 Features identified in Strip Area B, 3.09‐2.97 m amsl, Operation A ............ 182 Feature 27 in Strip Area C, Operation A ....................................................... 185 Features identified in Strip Area D and TP 60, 2.95‐2.90 m amsl, Operation A ................................................................................................... 186 Features identified in Strip Area D, 2.88‐2.80 m amsl, Operation A ............ 187 Features identified in Strip Area D, 2.71‐2.65 m amsl, Operation A ............ 188 Features identified in Strip Area D and TP 59, 2.47‐2.29 m amsl, Operation A ................................................................................................... 189 Representative plan views of Feature 44 at various elevations ................... 191 Strip Area E at an elevation of 2.84 m showing the location of Feature 46 .................................................................................................................. 195 Strip Area E at an elevation of 2.49 m showing the location of Feature 47 .................................................................................................................. 196 Strip Area G at an elevation of 2.74 m showing the location of Feature 32 .................................................................................................................. 197 Features identified in Strip Area H at 3.34 m (top) and 2.86‐2.74 m (bottom) amsl, Operation A .......................................................................... 198 Features identified in Strip Area Ja, 3.20‐3.09 m amsl, Operation A ........... 200 Features identified in Strip Area Ja, 2.98‐2.83 m amsl, Operation A ........... 201 Features identified in Strip Area Ja, 2.62‐2.41 m amsl, Operation A ........... 202 Features identified in Strip Area Ja, 2.12‐1.86 m amsl, Operation A ........... 203 Reproduction of PTA’s 1998 feature map showing the locations of Features 1, 2, 3, and Post Hole 1 (adapted from Campbell et al. 1998:Figure 24) ............................................................................................. 205 Profiles of postmolds/possible postmolds associated with a small structure in Strip Area Ja, Operation A ......................................................... 207 SEARCH excavation block in Strip Area Ja, Operation A. Bottom illustration shows unit numbers with shaded squares indicating excavated units. Top illustration shows the relationship of the backhoe trench and features to the block of units ..................................................... 208 Figure 6.35. View to the east‐northeast of feature excavations in Strip Area Ja, Operation A ................................................................................................... 210 Figure 6.36. “East” profile of Feature 106A ...................................................................... 210 x
Figure 6.37. Figure 6.38. Figure 6.39. Figure 6.40. Figure 6.41. Figure 6.42. Figure 6.43. Figure 6.44. Figure 6.45. Figure 6.46. Figure 6.47. Figure 6.48. Figure 6.49. Figure 6.50. Figure 6.51. Figure 6.52. Figure 6.53. Figure 6.54. Figure 6.55. Figure 6.56. Figure 6.57. Figure 6.58 Figure 6.59. Figure 6.60. Figure 6.61. Figure 6.62.
View to the southeast of Features 69/104, 120, and 64 .............................. 211 View to the south‐southwest of Features 64 and 120 ................................. 212 Profiles of Features 69/104 and 120: top) “north” wall of TPs 65, 70, and 71 across the backhoe trench; bottom) “west” wall of TPs 74, 75, and 76 . 212 Profile of Feature 120 showing layer of baked‐clay objects ......................... 214 Features identified in Strip Area Jb North, Operation A ............................... 216 Features identified in Strip Area Jb South, Operation A ............................... 217 View to the northeast of the top of Feature 60 after exposure in Strip Area Jb North, Operation A ........................................................................... 218 Profile of Feature 60 showing various depositional lenses. Compare with drawn profile in Figure 6.41 .......................................................................... 218 “West” profile of sample test in Feature 60. Arrow points to Lens 5 pocket that contained an incised sherd ........................................................ 219 Features identified in Strip Area K, 3.23‐2.94 m, Operation A ..................... 222 Features identified in Strip Area K, 2.25‐2.19 m, Operation A ..................... 223 North, west, and south wall profiles of PTA’s ST162 showing Feature 5. Reproduced from Meyer et al. 2002:Figure 20 ............................................ 224 Plan view of PTA’s TP 19 showing Features 6 and 7. Reproduced from Meyer et al. 2002:Figure 21 .......................................................................... 225 Features identified in Strip Area L and L‐North, 2.70‐2.57 m, Operation A .................................................................................................................... 226 Features identified in Strip Area L and L‐North, 2.28‐1.94 m, Operation A .................................................................................................................... 227 Profiles of postmolds and possible postmolds in Strip Area L, Operation A .................................................................................................................... 228 Features identified in Strip Area Q, Operation A .......................................... 231 Progression of work in Strip Area Y showing expansions and interior trenches ........................................................................................................ 232 Features identified in Strip Area Y at elevations 3.34‐2.98 m, Operation A .................................................................................................................... 233 Features identified in Strip Area Y, 2.89‐2.70 m amsl, Operation A ............. 234 Features identified in Strip Area Y, 2.83‐2.70 m and 2.20‐2.16 m amsl, Operation A ................................................................................................... 235 “West” profile of Feature 137, Strip Area Y, Operation A ............................ 237 South wall profile of Trench 4, Strip area Y, Operation A showing the relationship between Features 143 and 129/142 ........................................ 238 View to the west‐northwest of disturbed (darker) sediments containing a concentration of ceramic sherds in Strip Area Y at elevation 3.08 m. Note evidence of disturbance in the “west” wall of Strip Area Y ................. 239 Plan drawing of F14 and F17(P) in TP47, Building 632, Operation A. 14A represents a concentration of charcoal and fired clay; 14B and 14C represent piles of fired clay. Reproduced from Campbell et al. (2009:Figure 26) ............................................................................................ 241 PTA’s TP20 exposed in Strip area M, Operation B ........................................ 242 xi
Figure 6.63. Figure 6.64. Figure 6.65. Figure 6.66. Figure 6.67. Figure 6.68. Figure 6.69. Figure 6.70. Figure 6.71. Figure 6.72. Figure 6.73. Figure 6.74. Figure 6.75. Figure 6.76. Figure 6.77. Figure 6.78. Figure 6.79. Figure 6.80.
Figure 6.81. Figure 6.82. xii
Features identified in Strip Area M, Operation B ......................................... 243 Features identified in Strip Area Pb, Operation B ........................................ 244 View of west wall of the water main trench north of Building 635 showing the black sandstone feature. Note the darker soil strata above and below the sandstone. Photograph reproduced from Eglin Air Force Base 2003 ... 245 “North” wall profile of Strip Area Wa and TP95 showing modern intrusive pit intercepting previously deposited fill layers. Inset shows possible track marks in the profile of TP95 .......................................................................... 246 “North” wall profile of Strip Area Wa, TP97, and TP95 showing approximate original location of sandstone “feature” and the extent of a modern intrusive pit ..................................................................................... 247 Mass spectrum for the cholestanol peak in Feature 69/104 (top) and the NIST mass spectrum for cholestanol (bottom) ............................................. 250 Size ranges of postmolds and postholes: top) maximum length; bottom) maximum width ............................................................................................ 252 Size and depth ranges of pit features: top) maximum length; bottom) maximum depth ............................................................................................ 254 Frequency distribution of features by elevation in meters amsl ................. 256 Relationship of ground surface elevations and elevations of features ........ 256 Vertical relationship of ground surface and features in the midden area of Operation A: top) ground surface and features; bottom) ground surface and calibrated median radiocarbon ages before present (BP) as calculated by CALIB6.01. Dashed lines indicate the range of elevations where 50% of the features were first identified .......................................... 257 Distribution of feature types and midden deposits in Operation A ............. 258 Scatterplot of pit volume versus elevation indicating no significant correlation between the two variables ........................................................ 259 Distribution of pits, possible pits, pits and/or postholes by volume, midden area, Operation A ............................................................................ 260 Structure 1, Strip Area Jb: a) postmold arrangement showing two concentric lines of posts; b) same arrangement but with postmolds containing charcoal highlighted .................................................................... 262 Structure 2, Strip Area L, Operation A .......................................................... 263 Topographic map of Meig’s Pasture site (8OK102) showing semi‐circular arrangement of midden deposits. Reproduced from Curren (1987:Figure 3) .............................................................................................. 264 Comparison of the cumulative percentage distributions of major faunal classes by MNI for midden samples (Strata IVa and IVb) and analyzed features ......................................................................................................... 268 Bivariate plot of richness and evenness values for invertebrate taxa in midden strata (IVa and IVb) and analyzed features ..................................... 271 Comparison of most common invertebrate taxa by weight (≥ 100 g total for all five features) in .64‐cm fractions with taxa from F4 .......................... 273
Figure 7.1. Figure 7.2. Figure 7.3. Figure 7.4. Figure 7.5. Figure 7.6. Figure 7.7. Figure 7.8. Figure 7.9. Figure 7.10. Figure 7.11. Figure 7.12. Figure 7.13. Figure 7.14. Figure 7.15. Figure 7.16. Figure 7.17. Figure 7.18. Figure 7.19. Figure 7.20. Figure 7.21. Figure 7.22. Figure 7.23. Figure 7.24. Figure 7.25.
Formation of ellipses by the intersection of a cone with planes inclined at various angles ............................................................................................... 294 Idealized examples of a spheriod (left), oblate ellipsoid (center), and prolate ellipsoid (right) ................................................................................. 295 Examples of grooved (left) and finger‐impressed (right) spheroid baked‐clay objects (Thomas and Campbell 1991:Figure 6) .......................... 295 Example of an oblate ellipsoid (after Small 1966:74) ................................... 296 Example of a prolate ellipsoid baked‐clay object (Thomas and Campbell 1991:Figure 6) ............................................................................................... 296 Examples of amorphous baked‐clay objects (after Small 1966:76) ............. 297 Example of tabular baked‐clay object (after Campbell et al. 998:Figure 29) ............................................................................................... 297 Summary of baked‐clay object forms ........................................................... 304 Front and back views of grooved spheroid from Feature 22 (FS 115) ......... 304 Examples of oblate ellipsoids: a) Feature 105, FS481; b) TP61, FS76; c) TP58, FS7 ................................................................................................... 307 Example of prolate ellipsoid, TP6, FS478_126 .............................................. 308 Fragments of cylindrical baked‐clay objects: a) Feature 44, FS261, side and top views; b) Feature 120, FS617, note central hole ............................. 308 Tabular objects that display one flat surface and one rough surface: top) Feature 44, FS237; bottom) Feature 105, FS485 .......................................... 307 Amorphous baked‐clay object with finger impressions from TP60, FS46 .... 310 Amorphous baked‐clay object showing palm and fingernail impressions from TP92, FS533 .......................................................................................... 310 Examples of amorphous baked‐clay objects: a) TP40, FS090028_332; b) TP7, FS 478_140; c) TP1, FS478_194 ............................................................ 311 Results of principal components analysis of baked‐clay objects and clay samples ......................................................................................................... 313 Results of principal components analysis showing 95% confidence ellipses around the data points .................................................................... 314 Comparison of Bayou Park and Choctawhatchee Bay baked‐clay object samples ......................................................................................................... 314 Spatial distribution of baked‐clay objects by weight in grams ..................... 315 Proveniences containing greater than one percent of the total weight of baked‐clay objects at 8OK898 ...................................................................... 317 Feature 120, a clay‐lined cooking or roasting pit, with baked‐clay objects in situ ............................................................................................................. 318 Baked‐clay griddle fragments from AR‐39, Rio Tanamá, Puerto Rico. Note the flat upper surfaces (left) and irregular bottom surfaces (right). Compare with Figure 7.14. Adapted from Pagán Jiménez 2008:Figure 11.1 ........................................................................................... 320 Sherd‐like baked‐clay objects from Feature 60 ............................................ 321 Seriation of baked‐clay objects from three sites in the Choctawhatchee Bay region ..................................................................................................... 330 xiii
Figure 8.1. Figure 8.2. Figure 9.1. Figure 9.2. Figure 9.3. Figure 10.1. Figure 10.2. Figure 10.3. Figure 10.4. Figure 10.5. Figure 10.6. Figure 10.7. Figure 10.8. Figure 10.9. Figure 10.10. Figure 10.11 Figure 10.12. Figure 10.13. Figure 10.14. Figure 10.15. Figure 10.16. Figure 10.17. Figure 10.18. Figure 10.19. Figure 10.20.
xiv
Phytolith and Starch diagram for a griddle fragment from Bayou Park, 8OK898 .......................................................................................................... 355 Micrographs of phytoliths and starch recovered from a possible griddle fragment, Bayou Park site (8OK909) ............................................................ 357 Examples of decorated sherds from 8OK898 ............................................... 362 Swift Creek Complicated Stamped sherd with weathered surface and pencil rubbing showing the stamped design ................................................ 364 Vessel reconstruction of two rim sherds from 8OK898 ............................... 368 Locations of various geological formations containing toolstone in relation to Bayou Park .................................................................................. 386 Distribution of major ocean currents within the Atlantic and Gulf of Mexico basins ................................................................................................ 391 Distribution of major ocean currents within the Caribbean and Gulf of Mexico ........................................................................................................... 391 Westo points from Bayou Park ..................................................................... 394 Left: Biface scavenging and reuse. Right: Examples of biface reuse at Bayou Park .................................................................................................... 394 Little Bear Creek points from Bayou Park ..................................................... 396 Miscellaneous stemmed bifaces ................................................................... 397 Miscellaneous stemmed bifaces ................................................................... 400 Unstemmed bifaces ...................................................................................... 403 Biface fragments ........................................................................................... 405 Recycled biface fragment, CR‐051‐11‐281 ................................................... 406 Microliths from Bayou Park .......................................................................... 408 Examples of utilized flakes ............................................................................ 411 Examples of cores ......................................................................................... 413 Histogram showing the variation in flake weight between Tallahatta Quartzite and chert ....................................................................................... 417 Cumulative percentage distributions of flake‐form categories, Operations A and B .......................................................................................................... 419 Comparison of cumulative percentage distributions of flake‐form categories, Strata IVa‐c and VI with experimentally produced waste flake assemblages .................................................................................................. 419 Comparison of flake‐size distributions, Operations A and B, with experimentally produced waste flake assemblages ..................................... 420 Comparison of flake‐size distributions, midden (Strata IVa‐c) and non‐midden (Stratum VI), with experimentally produced waste flake assemblages .................................................................................................. 421 Comparison of flake‐size distribution from Bayou Park (midden and non‐ midden combined) with other archaeological assemblages demonstrating effect of differential access to raw materials on flake size .......................... 422
Figure 10.21. Figure 10.22. Figure 10.23. Figure 10.24.
Examples of cobble tools .............................................................................. 425 Examples of cobble tools .............................................................................. 426 Non‐cobble ground stone tools .................................................................... 429 Vertical distribution of lithic artifacts in Stratum VI in relation to feature depths in Operation A, Strip Area Q, and Operations B and C ..................... 438 Cumulative distribution of artifact classes in unmixed strata and features ......................................................................................................... 438 Lithic density distribution (artifacts per square meter) in shovel tests and test pits at Bayou Park .................................................................................. 439 Distribution of lithic artifacts by depth ......................................................... 441 Spatial distribution of lithic tools in Operation A ......................................... 442 Spatial distribution of lithic tool in Operation B ........................................... 443 Distribution of unmodified stone at Bayou Park .......................................... 444 Comparison of richness and evenness values for lithic tool assemblages from selected sites ........................................................................................ 450 Vertical distribution of stemmed bifaces ..................................................... 452 Hypothetical models of resource extraction for Bayou Park ....................... 456
Figure 10.25. Figure 10.26. Figure 10.27. Figure 10.28. Figure 10.29. Figure 10.30. Figure 10.31. Figure 10.32. Figure 10.33. Figure 11.1. Awls and a possible awl or dagger from Bayou Park .................................... 495 Figure 11.2. Spatula and pins from Bayou Park ................................................................ 497 Figure 11.3. Pointed fragments and a possible projectile point fragment from Bayou Park ............................................................................................................... 498 Figure 11.4. Antler artifacts from Bayou Park .................................................................. 499 Figure 11.5. Type E cutting‐edged tool (FS 785) from Strip Area Y .................................. 501 Figure 11.6. Method of hafting a Type E cutting‐edged tool ........................................... 501 Figure 11.7. Type F cutting‐edged tool (FS 786) from Strip Area Y, front and back views ............................................................................................................. 502 Figure 11.8. Columella hammers (a‐b) and a possible awl or dagger (c): a) FS 460, CS3, Strip Area L, Stratum VI; c) FS 305, Strip Area E, Stratum VI ................ 503 Figure 11.9. Back and front views of an awl or perforator, FS 247, F48 .......................... 504 Figure 11.10. Lightning whelk cup, FS 171, F36 .................................................................. 504 Figure 11.11. Gastropod shell dippers ................................................................................ 505 Figure 11.12. Shell debitage ................................................................................................ 506 Figure 12.1. Comparison of proportional representation of biomass by faunal class in midden column samples and features ......................................................... 548 Figure 12.2. Bivariate plot of ubiquity versus MNI ranking for edible taxa ...................... 551 Figure 12.3. Bivariate plot of ubiquity versus biomass ranking for edible taxa ............... 551 Figure 12.4. Comparison of exploited habitats ................................................................. 555 Figure 12.5. Comparison of canid tooth sizes ................................................................... 560 Figure 13.1. Radial cross‐section of a hard clam (Mercenaria spp.) exposing light (white) and dark (gray) incremental growth structures ............................... 606 xv
Figure 13.2. A backlit (A) and direct light (B) radial thin section of a hard clam shell (Mercenaria spp.) showing the translucent and opaque shell growth increments .................................................................................................... 607 Figure 13.3. Oxygen Isotope δ18O profile (A) showing the annual change in the concentration of 18O/16O in the shell of a hard clam (Mercenaria campechiensis) collected live in 1988 from Cedar Key, Florida .................... 608 Figure 13.4. Six‐part subdivision of translucent and opaque shell growth in hard clam (Mercenaria spp.) shells ................................................................................ 610 Figure13.5. Percentage of monthly hard clam (Mercenaria campechiensis) translucent and opaque shell growth ........................................................... 611 Figure 13.6. Seasonal and annual growth frequency profiles of hard clams (Mercenaria campechiensis) collected live 1987‐88 from Cedar Key, Florida ........................................................................................................... 612 Figure13.7. Growth frequency profile of hard clams (Mercenaria spp.) in pooled sample from Bayou Park ............................................................................... 613 Figure13.8. Growth frequency profile of hard clams (Mercenaria spp.) in Feature 36 from Bayou Park ........................................................................................... 614 Figure 15.1. Comparison of radiocarbon‐dated occupations at Bayou Park with Balsillie and Donoghue’s (2004) “Younger B” sea level model..................... 631 Figure 15.2. Artist’s reconstruction of a closed circular structure (top) and an open rectangular structure (bottom) similar to those identified at 8OK898 ........ 633 Figure 15.3. Examples of types of earth ovens ................................................................. 634 Figure 15.4. Bivariate graph comparing measures of residential stability and use‐duration for Bayou Park (8OK898) and Montverde (8LA243) ............... 641 Figure 15.5. Occupational chronology of Bayou Park based on radiocarbon dates and artifacts recovered from archaeological projects conducted between 1993 and 2012 .............................................................................................. 646 Figure 15.6. Comparisons of various stemmed bifaces identified as Destin points with Bullen’s Levy point and Haisten’s Crooked Creek point ............................... 658 Figure 15.7. Contour map of the Buck Bayou Mound (8WL90) showing the location of the test pit excavated by New World Research ....................................... 661 Figure 15.8. Distribution of Late Archaic sites in the Choctawhatchee Bay region ......... 666 Figure 15.9. Radiocarbon dates in calibrated years BP (2 σ) for sites with Elliott’s Point components .................................................................................................. 668 Figure 15.10. Radiocarbon dates in calibrated years BC (2 σ) for sites with Elliott’s Point components .................................................................................................. 669 Figure 15.11. Proposed interaction spheres and trade routes 4500‐3000 BP ................... 675 Figure 15.12. Distribution of steatite vessels and proposed exchange routes in Florida .. 676 Figure 15.13. Distribution of galena and proposed exchange routes in Florida ................ 676
xvi
LIST OF TABLES Table 1.1. Table 2.1. Table 2.2. Table 2.3. Table 3.1. Table 3.2. Table 3.3. Table 3.4. Table 3.5. Table 4.1. Table 4.2. Table 4.3. Table 4.4. Table 4.5. Table 4.6. Table 4.7. Table 4.8. Table 6.1. Table 6.2. Table 6.3. Table 6.4. Table 6.5. Table 6.6. Table 6.7. Table 6.8. Table 6.9. Table 6.10. Table 6.11. Table 6.12.
Summary of radiocarbon dates from 8OK898 ................................................ 23 Native American culture sequence in the Choctawhatchee Bay region ........ 50 Elliott’s Point chronology and diagnostic traits .............................................. 53 Calibrated radiocarbon dates for Elliott’s Point sites ..................................... 54 Characteristic traits of the Elliott’s Point Complex ......................................... 71 Characteristics of Bayou Park site ................................................................... 72 GPR grid sizes and location data ..................................................................... 80 Size and volume data for mechanical strip areas ........................................... 90 Grid coordinates and elevation data for SEARCH shovel tests and test pits .. 91 Strata correlations between representative shovel tests and test pits, midden area .................................................................................................. 110 Characteristics of soil samples from TP 60 ................................................... 120 Radiocarbon dates from Bayou Park ............................................................ 124 Proveniences of ceramic types from Bayou Park ......................................... 127 Proveniences of ceramic types recovered during PTA surveys and excavation ..................................................................................................... 130 Vertical distribution of stemmed bifaces in Operation A ............................. 134 Vertical distribution of stemmed bifaces in Operations B and C ................. 135 Archaeological components represented at Bayou Park ............................. 136 Summary of feature types at Bayou Park ..................................................... 174 Cultural materials recovered from sample unit, Feature 60 ........................ 220 Initial and dry weights of feature sediment samples and control samples.. 247 Lipid fractions ................................................................................................ 248 C27 isomer standards for qualitative identification of sample peaks .......... 248 Results of lipid residue analysis .................................................................... 249 Morphological characteristics of pit features ............................................... 253 MNI by faunal classes for midden strata (IVa and IVb) and analyzed features ......................................................................................................... 268 List of most common invertebrates (≥ 1% of total MNI) in midden strata (IVa and IVb) and analyzed features. The two most common taxa in each sample are indicated by bold type ................................................................ 269 Richness (R) and evenness (E) values for midden strata (IVa and IVb) and features ......................................................................................................... 271 Vertebrate remains from F4. Data from Scott 2002:Table 9 ....................... 272 Invertebrate remains from F4. Data from Meyer et al. 2002: Table 10 ...... 272
xvii
Table 7.1. Table 7.2. Table 7.3. Table 7.4. Table 7.5. Table 7.6. Table 7.7. Table 7.8. Table 7.9. Table 7.10. Table 7.11. Table 7.12. Table 7.13. Table 7.14. Table 9.1. Table 10.1. Table 10.2. Table 10.3. Table 10.4. Table 10.5. Table 10.6. Table 10.7. Table 10.8. Table 10.9. Table 10.10. Table 10.11. Table 10.13. Table 10.14. Table 10.15. Table 10.16. Table 10.17.
Baked‐clay objects recovered from 2011‐2012 Phase III excavations at 8OK898 .......................................................................................................... 299 Baked‐clay objects recovered during previous investigations at 8OK898 ... 301 Primary aplastic inclusions ............................................................................ 302 Secondary aplastic inclusions ....................................................................... 302 Colors represented in the baked‐clay sample .............................................. 303 Baked‐clay object forms and surface treatments ......................................... 304 Provenience data for baked‐clay object forms ............................................. 305 Results of pXRF analysis in parts per million (ppm) ...................................... 312 Baked‐clay objects that display evidence of post‐firing exposure to heat ... 318 Radiocarbon dates associated with baked‐clay objects ............................... 324 Stratigraphic data for baked‐clay objects at 8OK898, 2011‐2012 excavation ..................................................................................................... 326 Stratigraphic data for baked‐clay objects at 8OK898, 1998 PTA block excavation ..................................................................................................... 327 Stratigraphic data for baked‐clay objects at 8WL1278, all units combined . 328 Stratigraphic data for baked‐clay objects at 8OK877, Block 2, all units combined ...................................................................................................... 329 Ceramic types from Bayou Park (8OK898) ................................................... 361
Lithic artifacts recovered during previous projects at 8OK898 .................... 376 Proportional representation of flake types .................................................. 377 Cross‐tabulation of raw material by artifact class for PTA projects at 8OK898 .......................................................................................................... 378 Sample sizes from test units used in 8OK898 debitage analysis .................. 379 Lithic raw materials by artifact type ............................................................. 387 Total artifacts recovered from Bayou Park, 1993‐2012 ............................... 392 Data on stemmed biface, 8OK898 ................................................................ 393 Data on unstemmed bifaces, 8OK898 .......................................................... 403 Data on biface fragments, 8OK898 ............................................................... 404 Data on microliths from 8OK898 .................................................................. 408 Data on utilized flakes from 8OK898 ............................................................ 410 Data on cores from 8OK898 .......................................................................... 413 Summary data for lithic waste flakes ............................................................ 415 Results of Student’s t tests comparing mean flake weight between Operation A and B and between unmixed strata ......................................... 416 Summary data for lithic waste flakes by raw material type ......................... 416 Results of Student’s t tests comparing mean flake weight between raw material types in Operations A and B ........................................................... 416 Table 10.18. Flake‐form data for Operations A and B by strata ........................................ 418 Table 10.19. Flake‐size data for Operations A and B and for midden (Strata IVa‐c) and non‐midden (Stratum VI) contexts ............................................................... 420 Table 10.20. Technological flake types by strata ............................................................... 423
xviii
Table 10.21. Table 10.22. Table 10.23. Table 10.24. Table 10.25. Table 10.26. Table 10.27. Table 10.28. Table 10.29. Table 10.30. Table 10.31. Table 10.32. Table 10.33. Table 10.34. Table 11.1. Table 11.2. Table 11.3. Table 12.1. Table 12.2. Table 12.3. Table 12.4. Table 12.5. Table 12.6. Table 12.7. Table 12.8. Table 12.9. Table 12.10. Table 12.11. Table 12.12. Table 12.13. Table 12.14. Table 12.15.
Waste flakes of non‐siliceous stone ............................................................. 423 Cross‐tabulation of platform preparation and platform condition .............. 424 Data on cobble tools from 8OK898 .............................................................. 426 Data on non‐cobble ground stone artifacts .................................................. 428 Distribution of unmodified sandstone by strata in Operations A and B ...... 431 Provenience data for unmodified stone (non‐sandstone) ........................... 431 Cross‐tabulation of functional classes by artifact types ............................... 433 Distribution of lithic artifact classes by strata .............................................. 435 Distribution of lithic artifacts by unmixed strata .......................................... 436 Vertical distribution of artifacts in Stratum VI by depth .............................. 437 Comparison of observed and expected frequencies of lithic artifacts by strata and features, and chi‐square results .................................................. 439 Comparison of the degree of retouch on tool forms from midden (Strata IVa‐c), non‐midden (Strata VI), features, and other site contexts ............... 446 Comparison of degree of retouch among several sites in chert rich and chert poor areas of central and south Florida .............................................. 447 Comparison of functional tool types between 8OK898 and a sample of sites from peninsular Florida ........................................................................ 449 Provenience data for organic artifacts ......................................................... 494 Cross‐tabulation of artifact function by species ........................................... 494 Shell artifacts recovered during PTA 1998 block excavation ........................ 508 Faunal samples from Bayou Park (8OK898) chosen for analysis .................. 518 Subsampled and extrapolated totals for the 1/8‐inch and 1/16‐inch fractions of shell hash from Features 36 and 37 .......................................... 520 Allometric constants used to calculate biomass (g) from bone weight (g) .. 521 List of taxa identified in analyzed samples ................................................... 523 Summary of faunal data by class for Feature 36 .......................................... 527 Summary of faunal data by class for Feature 37 .......................................... 529 Summary of faunal data by class for Feature 44 .......................................... 530 Summary of faunal data by class for general levels in Feature 44 ............... 531 Summary of faunal data by class for Feature 46 .......................................... 532 Faunal classes represented in bulk samples from four features separated by screen‐size fractions ................................................................................. 534 Combined faunal data for Features 36, 37, 44, and 46 bulk samples separated by screen‐size fractions ............................................................... 537 Fauna recovered from Feature 60 separated by provenience and recovery mesh size ........................................................................................ 538 Summary of faunal data by class for Feature 60, all proveniences combined ...................................................................................................... 539 Fauna recovered from Feature 139 separated by levels .............................. 540 Summary of faunal data by class for Feature 139, all proveniences combined ...................................................................................................... 540 xix
Table 12.16. Table 12.17. Table 12.18. Table 12.19. Table 12.20.
Fauna recovered from CS1 and CS2 separated by stratum and level .......... 542 Summary of faunal data by class for Column Samples 1 and 2 combined ... 542 Fauna recovered from TP60, general excavation levels, separated by stratum and level .......................................................................................... 544 Summary of vertebrate faunal data by class for TP60, all general excavation levels combined .......................................................................... 544 Comparison of Biomass and MNI between midden column samples and features ......................................................................................................... 547 Ranking of edible taxa by ubiquity and abundance ...................................... 549 Comparison of burned fauna by class........................................................... 552 Comparison of burned fauna between midden column samples and features ......................................................................................................... 552 Comparison of exploited habitats based on fauna recovered from Bayou Park ............................................................................................................... 554 Comparison of fauna and exploited habitats between Bayou Park, Mitchell River 1, and Meig’s Pasture ............................................................ 562
Table 12.21. Table 12.22. Table 12.23. Table 12.24. Table 12.25. Table 13.1. Summary data on incremental growth phases in the shells of hard clams (Mercenaria spp.) from 8OK898 ................................................................... 613 Table 14.1. Summary of macrobotanical remains from 8OK898 .................................... 621 Table 14.2. Summary of macrobotanical remains from Feature 26 ............................... 621 Table 14.3. Summary of macrobotanical remains from Features 36, 37, 44, and 45 ..... 622 Table 14.4. Summary of macrobotanical remains from Feature 60 ............................... 623 Table 14.5. Summary of macrobotanical remains from Feature 106 ............................. 623 Table 14.6. Summary of macrobotanical remains from Feature 22 ............................... 623 Table 14.7. Summary of macrobotanical remains from Feature 105 ............................. 624 Table 14.8. Summary of macrobotanical remains from Feature 120 ............................. 624 Table 14.9. Summary of macrobotanical remains from Features 38 and 51 .................. 625 Table 14.10. Summary of macrobotanical remains from all samples ............................... 626 Table 15.1. Seasons of occupation based on various faunal and botanical indicators ... 637 Table 15.2. Comparison of feature types at Bayou Park, 8OK898, and Montverde, 8LA243........................................................................................................... 641 Table 15.3. Data used in the calculation of residential stability and use‐duration indices ........................................................................................................... 642 Table 15.4. Radiocarbon dates for fiber‐tempered sites east of Choctawhatchee Bay . 670 Table 15.5. Comparison of the proportional representation of baked‐clay object varieties by time period ................................................................................ 671
xx
LIST OF ACRONYMS AFB amsl ARPA BP cal cmbs DHR DoD EPO FAC FS FR GIS GPR GPS HPP ICRMP MNI NAGPRA NEPA NHPA NISP NPS NRCS NRHP NWR PRI PTA SEARCH SHPO USDA
Air Force Base above mean sea level Archaeological Resources Protection Act Before Present calibrated centimeters below surface Division of Historical Resources Department of Defense Elliott’s Point Object Florida Administration Code Field Sample Federal Register Geographic Information Systems Ground Penetrating Radar Global Positioning System Historic Preservation Plan Integrated Cultural Resources Management Plan Minimum Number of Individuals Native American Grave Protection and Repatriation Act National Environmental Policy Act National Historic Preservation Act Number of Individual Specimens National Park Service Natural Resource Conservation Service National Register of Historic Places New World Research, Inc. Paleo Research Institute, Inc. Prentice Thomas & Associates, Inc. Southeastern Archaeological Research, Inc. State Historic Preservation Officer United States Department of Agriculture
xxi
This page intentionally left blank.
xxii
ACKNOWLEDGMENTS SEARCH would like to acknowledge Mark Stanley, Lynn Shreve, and Joe Meyer from the 96th Civil Engineer Group Cultural Resources Section, Eglin Air Force Base, for their assistance throughout this project. Additionally, Cat Nolan of the Cultural Resources Section provided assistance with access to previously excavated collections and George Cole, also with Eglin’s Cultural Resources Section, provided previous data and maps, as well as guidance regarding use of the Eglin artifact database. Christopher Serbe Eglin AFB Contract Specialist was especially helpful in getting access for contractors and staff, sometimes on short notice. Joel Spaulding with the Real Property Office obtained access for SEARCH to use Building 584 for a field office and lab. Paul Steffen, SABER project manager, was a regular visitor to the site, and while anxious to move forward with the final demolition of Buildings 588 and 591, was enthusiastic about and interested in the archaeology of the site. SEARCH also wishes to thank several professional colleagues who offered assistance, data, and advice during the project. Janice Campbell with Prentice Thomas and Associates, Inc. (PTA) graciously provided copies of PTA reports and answered questions regarding past PTA excavation efforts at Bayou Park. Nancy White of the University of South Florida (USF) Anthropology Department assisted in the identification of ceramic sherds from the site, provided copies of articles, and offered advice and encouragement. Tom Pluckhahn of USF also helped in the identification of ceramic sherds. Mike Russo of the National Park Service’s Southeastern Archaeological Center provided references and advice on the fauna. Robert Tykot of USF performed the X‐Ray Fluorescence Spectrometry analysis of a sample of baked‐clay objects and USF doctoral candidate Rheta Lanehart conducted the lipid residue analysis of sediments from selected features. Harley Means with the Florida Bureau of Geology answered questions about the regional geology. Specialized analyses of the sclerochronology of hard clams was performed by Irv Quitmyer of the Florida Museum of Natural History, archaeobotanical analysis was performed Anna Dixon of USF, and phytolith and starch analysis of a griddle fragment was performed by Chad Yost of PaleoResearch, Inc. Radiocarbon dating was performed by Beta analytic, Inc. Finally, we would like to thank Al Miller with Geomill, Inc. for his skillful operation of the hydraulic excavator and for participating enthusiastically throughout the field work portion of this project. It was a pleasure to work with all of these individuals and institutions and to be able to share their commitment to and appreciation of the prehistory of Eglin Air Force Base.
xxiii
SEARCH PROJECT PARTICIPANTS Field Team Robert Austin Josh Torres Chris Mickwee Chris Altes Kendra Crain Josh Foster Linda Geary Matt Hanks Rockie Jarvis Tom Peterson Jon Simon Suarez Dennis Wardlaw Laboratory Analysis Debra Wells Jon Simon Suarez Robert Austin Travis Colliette Josh Torres Brian Worthington
Report Preparers Robert Austin Chris Mickwee Josh Torres Dennis Wardlaw Brian Worthington GIS Mapping, Photography, Graphics Curtis Deily Robert Austin Linda Geary Peer Review and Technical Editing Lisabeth Carlson Jennifer Salo Public Outreach Emily Powlen Meg Gaillard Project Management Anne Stokes James Pochurek Debbie Lowry
xxiv
EXECUTIVE SUMMARY Cultural Resources Data Recovery was conducted at the Bayou Park archaeological site (aka Former Eglin Prison Camp site), 8OK898, located on Eglin Air Force Base (Eglin AFB), Okaloosa County, Florida by Southeastern Archaeological Research, Inc. (SEARCH). The project was conducted according to the Scope of Work (SOW) included under Contract No. EAFB‐62221‐008, Task Order No. CR‐11‐0051. The purpose of this contract was to conduct archaeological data recovery at the Bayou Park site, a prehistoric archaeological site that has been determined eligible for listing in the National Register of Historic Places (NRHP). The work was carried out in fulfillment of responsibilities to consider the effects of construction activities on this NRHP‐eligible archaeological site. The Mid‐Bay Bridge Authority was the contracting authority for this project and Eglin AFB provided technical oversight. Mandating legislation and directives include the NHPA of 1966, as amended; the National Environmental Policy Act of 1969 (NEPA); 36 CFR Part 800; the Archaeological Resources Protection Act of 1979 (ARPA); the Native American Graves Protection and Repatriation Act of 1990 (NAGPRA); Air Force Instruction 32‐7065; and all laws, regulations and guidelines promulgated by the State of Florida governing cultural resources work. The background research for this project involved a review of all previous work conducted at 8OK898 as well as literature regarding the Elliott’s Point Complex in Florida and the Late Archaic generally. Copies of previous cultural resource studies at the site, as well as copies of original field notes and maps, were provided by the Eglin AFB’s Office of Cultural Resources. These also were used to obtain location information for all previous shovel tests, test pits, block excavations, and features that had been previously documented at 8OK898. Additional reports were obtained from the Florida Master Site File in Tallahassee, the SEARCH library in Newberry, and the libraries at the University of Florida and University of South Florida. In addition to a review of the literature, the artifact collections from previous surveys and excavations curated at Eglin AFB were examined and a sample of baked clay objects and lithic artifacts was reanalyzed. A ground‐penetrating radar (GPR) survey of the site was conducted in an effort to identify subsurface midden deposits, possible features, modern utility lines, and other subsurface anomalies. The GPR was successful in identifying the larger intact midden deposits as well as some of the larger subsurface features; however, the majority of features were not identifiable. Once the GPR survey was completed, 1‐x‐2‐m test units were excavated within the footprints of two demolished buildings, 588 and 591, followed by mechanical stripping within and surrounding these buildings. Work began here because of the need to remove footers and debris from the buildings. Once these buildings were cleared for debris removal, mechanical stripping proceeded to the east across the courtyard. Mechanical stripping was performed in 30 separate areas within 8OK898 for a total volume of approximately 4670 cubic meters. A total of 30 test excavation units were excavated: 11 1‐
x‐1‐m, 10 1‐x‐2‐m, 8 2‐x‐2‐m, and 1 .5‐x‐1‐m. The total volume of excavated sediments from these units was 74.86 cubic meters. Three column samples (.5‐x‐.5 m) also were excavated and soil samples from features and strata were obtained for laboratory analysis. Finally, 131 features were identified and mapped, including 112 prehistoric features, and a representative sample was excavated. The excavations resulted in the recovery of over 13,000 artifacts and over 500,000 faunal remains. The field work was considered complete when a level of data redundancy was reached such that additional excavation was considered unnecessary for the goal of mitigating adverse effect. Artifact processing and initial sorting was accomplished in the field. Final sorting and detailed analysis was conducted at the SEARCH laboratory in Newberry and by various specialists. Analyses continued through January 2013. In brief, the results indicate that the Bayou Park site was occupied primarily during the Late Archaic period 4481‐3766 cal BP (2529‐1817 cal BC), with minor occupations occurring during the Middle Archaic, Middle to Late Woodland, and Mississippi periods. The site consists of a semi‐circular midden deposit situated on a slight topographic rise in the southwest portion of the site and a moderately dense lithic scatter in the far eastern portion of the site near Postl Lake. In between these two areas is a relatively low‐density scatter of lithic artifacts, primarily waste flakes. The midden covers an area of approximately 1500 square meters and overlooks a wetland from which a small stream originates. The midden consists of dark, organic‐stained sediments containing animal bone, marine shell, and artifacts. Shell concentrations are evident within the darker midden matrix representing individual episodes of shell disposal which eventually began to overlap and coalesce in some parts of the site. The semi‐circular shape is the result of the structured organization of living space around a central “plaza” or common area where several very large pit features and a large clay‐lined hearth and earth oven were identified. Other large pits are located around the perimeter of the midden. Smaller pit features are more randomly distributed and are located within or immediately below the midden deposits. Some pits were clearly used for refuse disposal while others may have originally been used for storage. Lipid analysis of sediment samples from five features revealed the presence of cholestanol and epicoprostanol, which results from the degradation of terrestrial and marine animal remains. Plant sterols diagnostic of hickory nuts also were identified. Several post molds and postholes were identified and mapped leading to the recognition of two possible structures. One oval‐shaped structure measures no more than 2 or 2.5 m in diameter, although it was not entirely exposed. The second structure is larger and rectangular in shape, with a large open front and an associated hearth. It measures about 7 m across and about 2 m deep. The small sizes of the structures and their relatively insubstantial construction suggests they were inhabited by small family groups. xxvi
The lithic concentration near Postl Lake consists of at least two high density areas separated by a moderate to low‐density scatter of artifacts. One high‐density area was excavated during previous investigations at 8OK898 and includes a large pit containing marine shell and lithic tools. Another high‐density area was identified by SEARCH and contains primarily lithic waste flakes. Five other features also were identified in this part of the site. These are mostly refuse pits of various sizes, although the function of two features could not be determined. A total of 16 radiocarbon dates (four obtained during earlier investigations) identified three temporal occupations while two other possible occupations were identified on the basis of stratigraphy and temporally diagnostic artifacts. The primary occupation was during the Late Archaic period with 14 dates having a maximum two‐sigma range of 4481‐3766 cal BP (2529‐1817 cal BC). Artifacts associated with these dates include a variety of stemmed bifaces, microliths, a large number of baked‐clay objects, and, quartz, quartzite, and sandstone abraders. Along with the radiocarbon dates, these artifacts place the site’s major occupation in the earliest stages of the Elliott’s Point phase of the Late Archaic Period. Two other radiocarbon samples are from large wood posts and suggest later occupations. One sample returned a radiocarbon age of 1990 ± 30, or 1997‐1878 cal BP (48 BC‐AD 72). This would situate the feature within the Late Early Woodland period, which corresponds with the Deptford culture in northwest Florida. The second post returned a radiocarbon age of 250 ± 30 BP. The intercept crosses the calibration curve in several locations ranging from modern (‐1 to 11 cal BP) to between 428 and 376 cal BP (cal AD 1522‐1574), with a 75% probability that the true age is between 428 and 270 cal BP, or cal AD 1522‐1680. If correct, this would date the feature to the Spanish Contact period. No artifacts representative of either period have been recovered from Bayou Park. However ceramic sherds diagnostic of the Weeden Island and Fort Walton archaeological cultures were recovered from disturbed contexts. A possible Middle Archaic occupation is suggested by the presence of lithic waste flakes in deeply buried (70+ cmbs) contexts. No temporally diagnostic artifacts were recovered from these depths, although Middle Archaic stemmed bifaces (Marion/Newnan, Putnam, Levy/Pickwick) were recovered from the upper levels of the site and from disturbed spoil. Faunal remains are dominated by marine resources, particularly oysters, scallops, quahog clams, and a variety of fish species, particularly jacks. Terrestrial resources include deer, rabbit, gray squirrel, raccoon, turkey, crow, ducks, alligator, turtles, gopher tortoise, salamander, and lizards. Eighty‐nine different taxa were identified in the 298,000+ vertebrate and invertebrate specimens that were analyzed. Plant remains were not numerous but they provide important dietary and seasonality information. Edible plant remains include hickory nut shells, acorn shells, an unidentified seed, and an unidentified fleshy fruit fragment. Pine was the primary wood use for fuel along with hickory and oak. Phytolith and starch grain analysis of a griddle fragment identified bottle gourd, hackberry, xxvii
possible coontie, as well as phytoliths derived from wild grasses, sedges, and from the bark of members of the custard apple family. The presence of bottle gourd, which is not native to the New World, suggests the possibility of limited horticulture. The composition of two features, as well as the presence of a centrally located earth oven, was interpreted to be the result of small‐scale feasting for the purposes of reinforcing social identity. A number of seasonality indicators were identified in the faunal (quahog clam, fish), botanical (hickory nuts and acorns), phytolith (hackberry), and starch grain (coontie) samples. While all seasons of the year are represented by these samples, there is a congruence of all indicators during the early spring through late fall months. The presence of non‐local stone such as Tallahatta Quartzite and cherts from the Marianna and Wrights Creek Quarry Clusters suggest that the people who inhabited Bayou Park traveled north perhaps during the winter months to exploit these toolstone resource areas in Southern Alabama and near the Florida‐Georgia state line. Local raw materials include quartz and quartzite cobbles for grinding stones, sandstone for abraders, and clay for the manufacture of baked‐clay objects used in dry‐heat cooking. Powdered ferrous oxide material was found on one piece of sandstone suggesting that the sandstone fragments, which contain abundant iron, were being processed for pigments or dyes. No intact burials were identified but two human teeth were recovered from a single feature. After consultation between Eglin AFB and NAGPRA tribal representatives, the teeth were reburied on site as requested by tribal representatives. Given the large area of the site that has been investigated over the years, the potential for intact burials is considered to be remote. However, if human remains are encountered during construction activities, then the provisions outlined in NAGPRA should be followed. The Bayou Park site has provided a unique opportunity to examine a coastal Late Archaic, Elliott’s Point Complex site in some detail. It has produced important information on the age of the deposits, subsistence, settlement layout, seasonality, and mobility patterns which should assist in a better understanding of Elliott’s Point in the Choctawhatchee Bay region. Based on the results of this Phase III data recovery project, it is the opinion of SEARCH that adverse effects to 8OK898 resulting from proposed construction activities have been mitigated.
xxviii
1
INTRODUCTION
This report presents the results of a Cultural Resources Data Recovery project for the Bayou Park archaeological site (aka Former Eglin Prison Camp site), 8OK898, located on Eglin Air Force Base (Eglin AFB), Okaloosa County, Florida. The project was conducted according to the Scope of Work (SOW) included under Contract No. EAFB‐62221‐008, Task Order No. CR‐ 11‐0051. The purpose of this contract was to conduct archaeological data recovery at the Bayou Park site, a prehistoric archaeological site that has been determined eligible for listing in the National Register of Historic Places (NRHP). The site occupies about 20 acres on the former Eglin Federal Prison Camp. Several previous archaeological investigations have been performed at the site, each in response to specific site development projects in compliance with Section 106 of the National Historic Preservation Act (NHPA). These investigations established that the site’s prehistoric component dates primarily to the Late Archaic Period and contains archaeological remains associated with the Elliott’s Point Complex. The prison was officially closed in 2006 and plans for the multi‐purpose use of the property involve demolition of some buildings, rehabilitation of others, and extensive installation of underground utilities, renovation, and land modification. These undertakings pose a cumulative threat to the integrity of 8OK898. These threats can no longer be met by small‐scale data recovery projects focused only on areas of proposed impact. Therefore, to reduce land use restrictions and free the former prison property from further cultural resources obligations, a program of data recovery was developed to mitigate any future adverse effects. The work was carried out in fulfillment of responsibilities to consider the effects of construction activities on a National Register‐eligible archaeological site. The Mid‐Bay Bridge Authority was the contracting authority for this project and Eglin AFB provided technical oversight. Mandating legislation and directives include the NHPA of 1966, as amended; the National Environmental Policy Act of 1969 (NEPA); 36 CFR Part 800; the Archaeological Resources Protection Act of 1979 (ARPA); the Native American Graves Protection and Repatriation Act of 1990 (NAGPRA); Air Force Instruction 32‐7065; and all laws, regulations and guidelines promulgated by the State of Florida governing cultural resources work. The Principal Investigator for this project was Robert J. Austin, PhD. Josh Torres, PhD, served as Project Archaeologist. Both of these archaeologists exceed the qualifications established in the Secretary of the Interior's Standards and Guidelines (48 FR 44716 [29 September 1983]). The field crew included ten additional SEARCH archaeologists: Chris Altes, Kendra Crain, Josh Foster, Linda Geary, Matt Hanks, Rockie Jarvis, Chris Mickwee,
Bayou Park, 8OK898
Tom Peterson, and Dennis Wardlaw. Jon Simon Suarez, MA, directed the field lab. Robert J. Austin was the primary author of this report with contributions by Josh Torres, Brian Worthington, MA, Dennis Wardlaw, MA, Chris Mickwee, MA, Jon Simon Suarez, MA, Anna Dixon, PhD, Chad Yost, and Irvy Quitmyer.
PROJECT LOCATION AND PRESENT CONDITION The Bayou Park site (8OK898) is located in Sections 19 and 30, T1S, R22W in Okaloosa County, Florida (Figure 1.1). It is situated on a point of land where Boggy Bayou and Choctawhatchee Bay meet near the north‐central edge of the bay. Weekley Bayou forms the northern boundary of the point. Site boundaries were established through extensive subsurface shovel testing and limited excavations that began in 1993 and have been performed intermittently over the succeeding 17 years. Except for a narrow finger that extends to the northeast and separates a trailer park from Postl Lake, the site is contained primarily within an area bounded by Flagler Road to the northwest, Inverness Road to the southwest and northeast, and a brackish water lagoon known as Postl Lake to the southeast. This is the location of the former Eglin Federal Prison Camp which contains a complex of buildings surrounding a central courtyard (Figure 1.2). Three of the buildings at the west end of the compound (581, 588, and 591) were demolished in 2010, although footers, foundations, and surface rubble were still present when field work began in October 2011. Vegetation consists primarily of landscaped lawns and ornamental plants, although a grove of live oak, laurel oak, hickory, and a few sand pine is present at the southeastern end of the site, between the complex and the lake. Numerous utility lines (gas, water, electric, fiber‐optics, telephone) crisscross the compound resulting in localized disturbances. In addition, buildings and activities associated with the prison camp have impacted portions of the site.
SUMMARY OF PREVIOUS INVESTIGATIONS
Surveys and Excavations, 1993‐2010 The Bayou Park site has been subjected to numerous cultural resource investigations since it was discovered by Prentice Thomas and Associates, Inc. (PTA) during a cultural resource assessment survey of the Eglin Federal Prison Camp in 1993 (Campbell and Meyer 1993). This section summarizes the major surveys and excavations conducted between 1993 and 2010. Several smaller monitoring projects have been conducted by PTA and Eglin AFB cultural resource personnel during this time, but these are not discussed in detail unless significant discoveries were made. The 1993 survey consisted of the excavation of 69 50‐x‐50‐cm shovel tests at 20‐m intervals across a 32‐acre project area (Figure 1.3). Cultural materials were concentrated in an area measuring 90 by 120 meters in the tree‐covered southeastern portion of the property. The recovery of a Duval or Flint Creek projectile point indicated occupation during the Late
2
Introduction
Figure 1.1. Location of 8OK898 within Eglin Air Force Base, Okaloosa County, Florida.
3
Figure 1.2. Aerial photograph showing 8OK898 site boundaries within the former Eglin Federal Prison Camp.
Bayou Park, 8OK898
4
Introduction
Figure 1.3. Map of 1993 shovel testing at 8OK898 (from Campbell and Meyer 1993).
5
Bayou Park, 8OK898
Archaic or Early Woodland periods and the site was recommended as potentially eligible for the NRHP because of the presence of intact subsurface deposits. The boundary of the site was extended for about 180 m along the northern shore of Postl Lake in 1994 during a survey of several tracts of land within Eglin AFB (Matthews et al. 1994) (Figure 1.4). Further expansion of the boundaries and initial NRHP evaluation of the site occurred in 1996 (Mikell et al. 1996) (Figure 1.5). During this project, nine 50‐x‐50‐cm shovel tests and three 1‐x‐1‐m test units were excavated in the southeastern portion of the prison property. The site boundary was expanded slightly to the north as a result of the shovel testing. Apparently, this area had been covered with fill since the original 1993 survey and test units found that an intact deposit buried beneath this fill. The primary deposit occurred at a depth of 30‐40 centimeters below surface (cmbs) and extended to about 70 cmbs. Artifacts consisted primarily of lithic waste flakes and a few tools; however, a burned bone fragment and a small amount of shell also were recovered. In its 1996 report, PTA concluded that the site is unusual for the area because of evidence of both horizontal and vertical patterning in the distribution of lithic materials. Horizontally, artifacts were reported to occur in discontiguous concentrations while vertically a consistent depth of deposition was observed between about 40 and 70 cmbs. PTA also considered the site to be a relatively intact single component Late Archaic campsite. For these reasons, PTA recommended that 8OK898 be considered eligible for listing on the NRHP (Mikell et al. 1996:46‐47). In 1997, PTA excavated 15 50‐x‐50‐cm shovel tests and two 1‐x‐1‐m test units just south of the Food Services Building (Building 606, formerly Building 50522) in the southern portion of the prison camp (Campbell and Mathews 1997) (Figure 1.6). Only 22 waste flakes and a uniface fragment were recovered and these were limited in their vertical and horizontal distributions. The largest block excavation at 8OK898 prior to the current data recovery project occurred in 1998 when PTA investigated a 30‐x‐30‐m area in the empty lot at the northwest corner of the prison camp where a structure had existed and two new structures were planned (Campbell et al. 1998). The initial field work effort consisted of the excavation of 21 50‐x‐ 50‐cm shovel tests, one of which was expanded to a 1‐x‐1‐m unit (Test Pit [TP] 6) (Figure 1.7). These tests were the first to encounter intact midden deposits and subsurface features, and it expanded the site boundaries to the northwest. This was followed by auger tests, mechanical stripping, and block excavation of 12 1‐x‐1‐m units (Figure 1.8). The midden was found to be buried at depths varying from 27 to 53 cmbs. It consisted of very dark grayish‐brown sand in some areas and a shell midden in other areas. The upper 7‐12 cm consisted of recent fill underlain by Ap and AC horizons which developed over the midden after the site was abandoned. The midden rests on the preoccupation ground surface (Ab3) and underlying C horizon. 6
Introduction
Figure 1.4. Map of 1994 shovel testing at 8OK898 (from Matthews et al. 1994).
Figure 1.5. Map of 1996 shovel testing and test pit excavation at 8OK898 (from Mikell et al. 1996).
7
Bayou Park, 8OK898
Figure 1.6. Map showing the location of the 1997 shovel testing and test pit excavation at 8OK898 (from Campbell and Matthews 1997). Note that Building 50522 was subsequently renumbered as 606.
Figure 1.7. Map showing the location of the 1998 shovel testing at 8OK898 (from Campbell et al. 1998). Note that Buildings 50501 and 50502 were subsequently renumbered as 591 and 588, respectively.
8
Introduction
Figure1.8. Map showing the location of the 1998 block excavation at 8OK898 (from Campbell et al. 1998).
9
Bayou Park, 8OK898
The features consisted of a linear shell concentration (Feature 1), a possible cooking pit (Feature 2), a possible refuse pit (Feature 3), and a postmold. Shell from Feature 1 was radiocarbon dated to 3560 ± 70 measured 14C years BP (2205‐1860 cal BC) while shell from Feature 2 was dated to 3790 ± 70 measured 14C years BP (2535‐2140 cal BC) (Morehead, Campbell, and Aubuchon 2008:59). The artifact assemblage included lithic debitage and a small number of tool forms (n=513), pieces of ground stone (n=10), unmodified sandstone and limonite (n=23), worked shell (n=41), fired clay (n=943), Elliott’s Point Objects (EPOs; n=63), and ceramic pottery sherds (n=17). The EPOs are described as primarily amorphous in shape, but some tabular and melon‐shaped forms were identified. Temporally diagnostic artifacts include one Westo projectile point, a Carrabelle Punctated sherd, a sherd of Cool Branch Incised, two brushed sherds, 12 plain sherds, and the Elliott’s Point Objects (EPOs). In addition to the Westo point, two microliths (identified as Jaketown perforators), a core, seven biface fragments, and a uniface were recovered. Most of the lithic material was identified as Tallahatta Quartzite with a small, but varied amount of cherts also present. The Westo point, Jaketown perforators, and EPOs were considered consistent with a Gulf Formational (Late Archaic)‐period, Elliott’s Point Complex site and the radiocarbon dates indicated that this part of 8OK898 was occupied primarily during the early phase of Thomas and Campbell’s (1993) Elliott’s Point chronology (see Chapter 2). The pottery sherds also indicate occupation, albeit non‐intensive, during the Weeden Island and Fort Walton/Pensacola periods. Faunal analysis was limited to a sample obtained from TP12 and three features. The dominant shellfish was oyster, while quahog clam, bay scallop, and lightening whelk were common and rangia clams were very scarce. Most of the shell from Feature 2 (99%) was burned, which led to the conclusion that this may have been a cooking pit. Vertebrate fauna included turtle, white‐tail deer, fish, opossum, and bird. Analysis of charred botanical remains identified pine as the dominant wood species followed by oak. Cherry, elm, white cedar, and grape vine were present in small amounts. Hickory nut shell was found in two features (Feature 1 and Feature 3) and the general midden leading the authors to conclude that the site was occupied during the fall months. The next significant cultural resource project occurred in 2002 when PTA conducted a survey of a proposed sprinkler and communication cable trench (Meyer et al. 2002) (Figure 1.9). Forty‐five 50‐x‐50‐cm shovel tests were excavated along the 620‐m‐long trench line at 20‐m intervals. The trench line began in the northwest corner of the prison camp south of Building 50501 (currently 591) and near the block excavation conducted in 1998. It then proceeded north to the compound fence, then southeast along the fence line to a point just past Building 50523 (currently 635) where it turned southwest and then west, terminating at Building 50525 (currently 636). The shovel testing identified two areas of possible occupation: in the northwest part of the compound near Buildings 50501 (591) and 50502 (588), and in the southeast near Building 50525 (636). 10
Introduction
Figure 1.9. Map showing the location of the 2002 shovel testing and test pit excavation at 8OK898 (from Meyer et al. 2002).
In the northwestern area, Shovel Test (ST) 163 encountered shell midden between 35 and 60 cmbs resting on a probable pre‐occupation Ab horizon. Shovel test 162 encountered a refuse pit between 48 and 150 cmbs. A 1‐x‐1‐m test pit (TP19) excavated near these shovel tests encountered the midden and identified two additional features at about 150 cmbs. The strata in TP19 were similar to those encountered during the 1998 block excavation to the northwest except that the fill was much thicker. The upper fill zone measured 50‐cm thick and overlay A and AC horizons, which formed over the top of the buried midden. The midden consisted of very dark gray sand from 108‐150+ cmbs. The excavation was terminated before the base of the midden was reached since the proposed trench would not extend that deep.
11
Bayou Park, 8OK898
No midden strata were identified in the southeastern portion of the compound; however, a fourth feature was identified in ST174. It was subsequently investigated with two 2‐x‐2‐m excavation units (collectively identified as TP20) and found to be a bell‐shaped refuse pit. The top of the feature was encountered at about 70 cmbs and extended to a depth of 145 cmbs. It contained primarily crushed oyster, quahog clam, and smaller amounts of Rangia, coquina, and scallop shells, bone fragments, EPOs, and lithics. The lithic artifacts included a Ledbetter‐like projectile point. Three other projectile points were recovered from outside the features: Six Mile Creek, Destin‐like, and an unidentified “dart point.” Two shell samples from the bell‐shaped feature returned radiocarbon dates of 3800 ± 50 measured 14C years BP (2490‐2180 cal BC) and 3770 ± 50 measured 14C years BP (2460‐2130 cal BC) (Beta Analytic, Inc. 2002). These dates are very similar to the dates obtained from Features 1 and 2 excavated in the northwest portion of the site (described above), suggesting that the two areas were relatively contemporaneous. In addition to the projectile points, other artifacts from this project included lithic debitage (n=247), bifaces (n=7), microliths (n=3), utilized or retouched flakes (n=3), pottery (n=1), EPOs (n=23), and fired clay (n=214). A small sample of botanical remains was collected from Feature 4, but the only species identified was a single hickory nut shell fragment. Vertebrate faunal remains included white‐tail deer, fox squirrel, dog or coyote, turkey, box turtle, mullet, red and black drum, and crevalle jack. Invertebrates included oyster, quahog clam, rangia, scallop, banded tulip, moon snail, crown conch, coquina, cross‐barred venus, cardita, ladder shell, common slipper, and nerite. During this project, PTA established a permanent grid system with the N1000, E1000 point located about six meters to the northeast of a concrete sewer marker in the northwest corner of the prison camp. The 1998 block excavation unit coordinates were revised to reflect the new grid and all subsequent work was also tied into this system. To facilitate this, secondary datum points were established in various locations throughout the site. In 2003, Eglin cultural resource personnel monitored excavation of the trench and identified a linear sandstone feature just to the northeast of Building 635 (formerly 50523). The function of this feature is not known, nor has it been determined to be definitely cultural (Eglin Air Force Base 2003). In 2007, work was undertaken to investigate areas that would be impacted by the installation of fiber‐optics cable within the prison compound (Morehead, Campbell, and Aubuchon 2008) (Figure 1.10). Eight 50‐x‐50‐cm shovel tests were excavated in various locations within the compound and 16 similar shovel tests were excavated in a proposed parking lot to the northeast of the compound. In addition to the shovel tests, one 2‐x‐2‐m unit (TP21) and seven 1‐x‐1‐m units (TPs 22‐28) were excavated in proposed impact areas. Three of the shovel tests (176, 177, and 183) and four of the test units (21, 23, 24, and 26) contained cultural materials. 12
Introduction
Figure 1.10. Map showing the location of the 2007 shovel testing and test pit excavation at 8OK898 (from Morehead, Campbell, and Aubuchon 2008).
While the shovel testing expanded the boundaries of the site slightly to the north and northeast, no new midden areas were identified. The 2‐x‐2‐m test unit (TP21) was excavated near the northern corner of Building 604. The upper 30 cm were disturbed, but lithic waste flakes were recovered from undisturbed sediments between 30 and 130 cmbs. Test Pit 23, located near the southwest corner of Building 604 displayed disturbed sediments to a depth of 60 cmbs with only a single lithic waste flake recovered from between 90 and 100 cmbs. Test Pit 24, located near the southeast corner of Building 626, was disturbed to 40 cmbs, but lithic waste flakes were recovered between 60 and 80 cmbs. In TP26, on the south side of Building 633, lithic and shell fragments were recovered from between 40 and 50 cmbs beneath an upper disturbed layer; however, no additional artifacts or shell were recovered below 50 cmbs and excavation ceased at 100 cmbs. Artifacts were not abundant; the total number from all test pits consisted of 22 waste flakes, 19 of which were identified as Tallahatta Quartzite and the remainder as Coastal Plain chert.
13
Bayou Park, 8OK898
PTA provided a map in this report showing the depth of fill across the prison compound based on the results of shovel tests and test units. This map shows that the depth of fill and disturbed sediments varies from 25 to 75 cmbs across the compound, with the deepest occurrences near existing structures. Subsequent to this investigation, the location of the fiber‐optic line near Buildings 604 and 633 were moved by the contractor and Eglin AFB cultural resource personnel excavated two test units within the new area of potential effect (APE) (Morehead et al. 2008: Appendix A). Test Pit 29 (1‐x‐.8 m) was excavated next to PTA’s TP26 south of Building 633 and one chunky fragment of Tallahatta Quartzite and an incised EPO fragment were recovered from between 50 and 60 cmbs. Test Pit 30 (1‐x‐1‐m) was excavated near PTA’s ST175 on the west side of Building 604. Three lithic chunks of Tallahatta Quartzite and Coastal Plain chert and one EPO were recovered from between 60 and 90 cmbs. In 2008, excavations were conducted to mitigate impact from the installation of a concrete refrigeration pad and an electrical utility line associated with Building 635 in the northeastern portion of the Prison compound (Morehead and Aubuchon 2008). Five 1‐x‐1‐ m units were judgmentally located where the concrete pad was proposed (Figure 1.11) and two 1‐x‐1‐m units and two 50‐x‐50‐cm shovel test (STs 199, 200) were excavated where the utility trench was planned (Figure 1.12). Both excavation areas exhibited rather extensive disturbance. In the concrete pad area, fill and disturbed sediments were encountered to depths ranging from 35 to 60 cmbs and utility trenches were found to have caused localized disturbance to depths of up to one meter. Within the proposed utility line, fill was encountered to depths ranging from 50 to 70 cmbs with utility trench disturbance to at least one meter. Cultural materials were recovered from three test units in the concrete pad area and were extremely limited in number consisting of seven Tallahatta Quartzite waste flakes and two pieces of shell. In 2009, PTA conducted monitoring of the construction of a parking lot, two driveways, sidewalks, and a retention pond to the northeast of the Prison compound, as well as improvements in the air conditioner area of Building 635 (Bourgeois et al. 2009). Portions of the parking lot and retention pond areas were carefully graded. Although historic artifacts were occasionally found and a historic refuse deposit was identified and tested (TP38) in the retention pond area, no prehistoric artifacts or features were documented. Nor were any artifacts or features identified in the air conditioner area near Building 635. Also in 2009, PTA conducted mitigative excavations associated with renovations to Buildings 632, 633, and 634 (Campbell, Morehead, et al. 2009). Utility trenches and water line trenches were planned outside the buildings while impacts to the interiors involved the removal of the wooden flooring and bases. Fourteen 50‐x‐50‐m shovel tests, one 1‐x‐1‐m test pit (TP49), and four 2‐x‐2‐m test pits (TPs 39‐42) were excavated in various proposed impact areas outside the buildings and 20 50‐x‐50‐cm shovel tests, three 1‐x‐2‐m test units, one 1.4‐x‐2‐m test unit, and two 2‐x‐2‐m test units were excavated within the interiors of 14
Figure 1.11. Map showing the location of the 2008 test pit excavation northeast of Building 635 at 8OK898 (from Morehead and Aubuchon 2008).
Figure 1.12. Map showing the location of the 2008 test pit excavation southwest of Building 635 at 8OK898 (from Morehead and Aubuchon 2008).
Introduction
15
Bayou Park, 8OK898
the three buildings after the flooring had been removed (Figure 1.13). Unlike previous excavations, the 2009 project utilized a datum plane for determining depths during unit excavation. The datum plane was based on a true elevation of 3.88 m above mean sea level (amsl) established at a manhole cover near the northwest corner of Building 633. Elevations were transferred to the northwest corners of test units and level measurements were recorded in cm below unit datum. Fill layers were removed by hand without screening and controlled excavation using 10‐cm levels began at the top of the exposed A or E horizons. Many of the exterior shovel tests and test pits encountered heavily disturbed sediments; however, a few encountered prehistoric artifacts in what appeared to be undisturbed contexts beneath the disturbed upper strata. Test Pit 49 (1‐x‐1‐m), located northeast of Building 633, penetrated disturbed sediments to 117 centimeters below site datum (cmbd), but recovered lithics, fired clay fragments and shell in presumably undisturbed contexts between 118 and 187 cmbd. Shovel Test 209 was excavated on the east side of Building 634. Although the upper 66 cm were disturbed, marine shell was recovered from what appeared to be undisturbed sediments between 104 and 114 cmbd. The authors remark on the variety of shellfish remains in this small sample, which included oyster, quahog clam, Rangia, scallop, and ark. Test Pit 39 was located on the northwest side of Building 632 and contained shell, lithics, and fired clay between 60 and 163 cmbd. Test Pit 40, placed on the northwest side of this building, also recovered shell, lithics, and a large quantity of fired clay below 60 cm of disturbed sediments. Prehistoric artifacts in this unit continued to be recovered to 142 cmbd. One flake was found between 119 and 129 cmbd in TP41, located on the northeast side of Building 633. Test Pit 42, placed on the southeast side of Building 634, was the least disturbed of all the exterior tests. Fill material was encountered only in the upper 30‐35 cm which overlay undisturbed sediments containing lithic waste flakes, fired clay, and shell. Excavation was terminated at 199 cmbd. Excavations within the structures encountered deep fill deposits of up to 65 cm in thickness and localized disturbances below this depth. In Building 634, for example, asphalt slabs were encountered at 95 cmbd and several historic features and postmolds were identified. Despite the fill and disturbance, three prehistoric features were identified in the deeper, undisturbed strata. Features 14 and 17 were identified in TP47 within Building 632. The two features blend into one another and are probably functionally related. Feature 14 was encountered at 146 cmbd and Feature 17 at 173 cmbd. Feature 14 is interpreted as a probable hearth and small piles of charcoal and fired clay were found nearby suggesting that these may have come from cleaning out the hearth. The third prehistoric feature was identified in TP44 within Building 634. Feature 16 was encountered at a depth of 168 cmbd and extended for about 20 cm. It was oblong in shape and contained charcoal flecks, two waste flakes, and a piece of fired clay. It was interpreted as a possible hearth. 16
Introduction
Figure 1.13. Map showing the location of the 2009 shovel testing and test pit excavation at 8OK898 (from Campbell, Morehead, et al. 2009).
17
Bayou Park, 8OK898
According to Campbell, Morehead, et al. (2009:84), the soils encountered during this project conform to the Resota series, with an A‐E‐(Bh)‐Bw‐C sequence of horizons beneath the modern fill episodes. The sequence typically displayed a dark Ap horizon when present; a white E horizon; a yellowish‐brown to brownish‐yellow Bw horizon; and a brownish‐ yellow C horizon. No midden‐stained sediments were encountered in any of the shovel tests or test pits excavated during this project. Most of the recovered artifacts were historic; prehistoric artifacts consisted of 96 waste flakes, 3 EPOs, and 2291 g of fired clay. One of the flakes is a small bladelet that resembles a possible blank for a microlithic tool. It was recovered from between 137 and 147 cmbd in TP49. Typical for this site, most of the debitage is Tallahatta Quartzite with smaller amounts of Coastal Plain chert, miscellaneous chert, metaquartzite, and quartz. The EPOs were found in two units, TPs 43 and 44, all below 154 cmbd. Invertebrate remains included primarily oyster and quahog clam as well as smaller amounts of scallop, ark, rangia, cross‐barred venus, and horn shell. Most of the shell is considered to be from prehistoric contexts. The vertebrate material, which included some domesticated species, appears to be primarily from historic contexts, although the authors indicate that at least some of the non‐domesticated species may be prehistoric. In 2010 data recovery efforts to mitigate impact from the restoration of Building 636 were conducted by PTA (Morehead et al. 2011). The renovation plans involved trenching on the exterior of the building for electrical and water lines, and the installation of two concrete pads. Twenty‐two 50‐x‐50‐cm shovel tests, two 1‐x‐.5‐m test pits, two 1‐x‐1‐m test pits, and four 1‐x‐2‐m test pits were excavated during the project (Figure 1.14). Most of the units encountered fill and disturbed sediments at thicknesses varying from 20 to 60 cm. Localized disturbances from utility trenches reached as deep as 90 cmbs. Artifacts included mostly lithic waste flakes and fragments of fired clay, although a biface fragment, microlithic perforator, and an amorphous EPO are reported. Lithics were recovered as deep as 110 cm and the EPO came from 70‐80 cm. Test Pits 54, 56, and 57 contained a dense concentration of lithics between 10 and 100 cm. Faunal material included a small amount of oyster shell. One hickory nut shell was recovered from 80‐90 cmbs in TP52. PTA also monitored the removal of interior floor slabs. Since the maximum depth of impact was only 25 cm, the underlying soil was turned over with a shovel and probed to determine if any midden deposits are present beneath the building. None were encountered and fill was identified to a depth of 40 cmbs. The most recent investigation at 8OK898 occurred in November 2010 when Eglin cultural resource personnel monitored the demolition of Buildings 581, 588, and 591 (Speal and Nolan 2011) (Figure 1.15). A sewer trench excavated to the north of Building 591 encountered dark midden soils beneath recent fill at a depth of about 75 cm. The midden contained a variety of shell species (scallop, oyster, quahog clam, lightning whelk, fighting conch, moon snail, ark, and rangia) as well as fish and mammal bones. One quartz flake was also collected. Two areas of midden shell were identified within Building 591. One of these 18
Introduction
Figure 1.14. Map showing the location of the 2010 shovel testing and test pit excavation at 8OK898 (from Morehead et al. 2011).
19
Bayou Park, 8OK898
Figure 1.15. Map showing the location of the 2010 monitoring and the area at Building 591 where midden deposits were identified (from Speal and Nolan 2011).
(designated Area A) was immediately south of the sewer trench while the second (Area B) was located near the west end of the building. The shell was contained within clay fill that underlay the building and is assumed to have become mixed when the ground surface was graded and prepared prior to building construction. These findings extended the northern and eastern edges of the buried midden in this part of the site.
Summary Between 1993 and 2010, 257 50‐x‐50‐cm shovel tests and 57 test units varying in size from 1‐x‐.5‐m to 2‐x‐2‐m in size were excavated at 8OK898, and several small‐scale monitoring projects were carried out. These projects produced a large body of data on site stratification and vertical and horizontal distributions of artifacts, faunal remains, and features, as well as chronological data that places the site within the earliest phase of the Late Archaic Elliott’s Point Complex. This work laid a strong foundation for the 2011‐2012 data recovery project to build on and strongly influenced the approach to excavating and analyzing the site. Figures 1.16 and 1.17 show the locations of all PTA shovel tests and test pits excavated between 1993 and 2010. 20
Figure 1.16. Map of shovel tests excavated by PTA, 1993‐2010.
Introduction
21
Figure 1.17. Map of test pits excavated by PTA, 1993‐2010.
Bayou Park, 8OK898
22
Introduction
Four radiocarbon dates on shell indicate that the site was occupied between 4210 and 3970 years BP (recalibrated here using CALIB 6.01 as 2529‐1817 BC) (Table 1.1). This places the site well within the Late Archaic period and the “nascent” Elliott’s Point complex as defined by Thomas and Campbell (1993). Moreover, there is very close correspondence between the four dates, particularly those from Feature 4, the bell‐shaped pit identified in southeast portion of the site, and Feature 2, identified in the midden area in the northwest corner of the site. Pairwise comparison using a student’s t test determined that these three dates are not significantly different from one another and yield a pooled average age of 4196 ± 36 BP. On the other hand, all three ages are significantly different from the conventional 14C age obtained from Feature 1. These dates suggest that 8OK898 may have been occupied over a period of about 200 years or so. Sample No.
Table 1.1. Summary of radiocarbon dates from 8OK898. Measured 13C/12C Conventional Calibrated Date Provenience 14 14 C Age Ratio C Age Range (2 sigma)
Beta‐171507
Feature 4
3800±50
0.0
4210±60
2529‐2178 B.C.
Beta‐129298
Feature 2
3790±70
0.0
4200±70
2532‐2135 B.C.
Beta‐171508
Feature 4
3770±50
0.0
4180±60
2468‐2132 B.C.
Beta‐129297 Feature 1 3560±70 0.0 3970±70 2228‐1817 B.C. NOTE: Calibrations performed using CALIB 6.01 and MARINE09 calibration dataset (Stuiver and Reimer 1993).
The main areas of artifact concentration appear to be the midden area in the northwestern corner of the site and an area of lithic artifacts, baked‐clay objects, and features to the southeast near Postl Lake. The midden area appears to be unique in having dark, organic‐ stained sediments, abundant faunal remains, dense concentrations of baked‐clay objects and fired clay fragments, and several features, including a postmold from a possible structure. Shovel testing and the excavation of a 1‐x‐1‐m unit (TP19) to the east of the 1998 block excavation indicate that the midden extends to the north in front of, and possibly underneath, Building 591. Additional shovel tests to the southeast of the sidewalk bordering the block excavation area also encountered midden material suggesting an extension in this direction as well. A second area of interest is the far eastern portion of the site near Postl Lake. PTA’s TP20, which actually includes two overlapping 2‐x‐2‐m units, was situated here and exposed a large refuse pit (Feature 4). In addition, TP20 contained a large assemblage of worked lithics, including four projectile points (one from Feature 4), three bifaces, two microliths, and two utilized flakes. Burned bone was recovered between 20 and 60 cmbs in TP3 located to the north of TP20. This area was the focus of a recent data recovery project by PTA (Morehead et al. 2011) and it identified a dense concentration of lithics in TPs 54, 56, and 57. A third area of interest is located in and around Buildings 626, 632, 633, 634. Although artifact density within this area is very low, marine shell was consistently recovered from
23
Bayou Park, 8OK898
test units placed underneath and near these buildings. Several pit features and a postmold were identified in TPs 44, 45, and 47. In addition, a linear sandstone feature of unknown age or function was identified during monitoring to the north of Building 635. Although testing in the rest of the site has been limited, there appears to be only very low densities of artifacts and minimal amounts of faunal material located in these areas and no features were identified.
SUMMARY OF PHASE III RESULTS
Excavation Strategy The Phase III data recovery project began on October 17, 2011 with a Ground‐Penetrating Radar (GPR) survey. Excavation began on November 28, 2011 and was completed on February 21, 2012. In order to organize the work effort and to coordinate the location of buried utilities, the site was subdivided into three arbitrary areas or Operations (Figure 1.18). Operation A included the midden area identified by PTA and extended about halfway into the central courtyard. Operation B included the area east of Buildings 635 and 636. Operation C included the courtyard area between Operations A and B. The first task was to conduct a GPR survey of the entire site in an effort to identify subsurface midden deposits, possible features, modern utility lines, and other subsurface anomalies. Once the GPR survey was completed, 1‐x‐2‐m excavation units were begun within the footprints of two demolished buildings, 588 and 591, followed by mechanical stripping within and surrounding these buildings. Work began here because of the need to remove footers and debris from the buildings. Once these buildings were cleared for debris removal, mechanical stripping proceeded to the east across the courtyard. Once stripping in Operation A was completed, work moved to Operation B, then to Operation C. Finally, some additional stripping was conducted in Operation A to the east of Building 591 after the building debris had been removed in order to further explore the extent of the midden. At the same time that the stripping took place, excavation of 1‐x‐2‐m and 2‐x‐2‐m test pits was conducted in selected areas. The test pits were excavated in areas where stripping was not feasible or to investigate areas of artifact or faunal concentrations identified during previous work by PTA. Mechanical stripping was performed in 30 separate areas within 8OK898 ‐‐ 19 in Operation A, 8 in Operation B, and 3 in Operation C – for a total volume of approximately 4670 cubic meters. A total of 30 test excavation units were excavated: 11 1‐x‐1‐m, 10 1‐x‐2‐m, 8 2‐x‐2‐ m, and 1 .5‐x‐1‐m. The total volume of excavated sediments from these units was 74.86 cubic meters. Most of the test units were excavated in Operation A (n=21), including 10 1‐ x‐1‐m units in a small block that was excavated to investigate several large, overlapping features. Four units were excavated in Operation B and five in Operation C. Three column samples (.5‐x‐.5 m) also were excavated and soil samples from features and strata were obtained for laboratory analysis. Finally, 131 features were identified and mapped,
24
Figure 1.18. Map showing Operations A, B, and C within site 8OK898 as well as the locations of Strip Areas and Test Units.
Introduction
25
Bayou Park, 8OK898
including 112 prehistoric features, and a representative sample was excavated. The excavations resulted in the recovery of nearly 13,000 artifacts and over 500,000 faunal remains.
Excavation Results The details of the excavation and various analyses are the subjects of the following chapters. In brief, the results indicate that the Bayou Park site was occupied primarily during the Late Archaic period 4481‐3766 cal BP (2529‐1817 cal BC), with minor occupations occurring during the Middle Archaic, Middle to Late Woodland, and Mississippi periods. The site consists of a semi‐circular midden deposit situated on a slight topographic rise in the southwest portion of the site and a moderately dense lithic scatter in the far eastern portion of the site near Postl Lake. In between these two areas is a relatively low‐density scatter of lithic artifacts, primarily waste flakes. The midden is estimated to cover an area of approximately 1500 square meters. It is oriented with its open side facing northwest, or upslope. The position of the midden at some distance from Postl Lake and Choctawhatchee Bay is believed to be due to its location on high ground overlooking a wetland from which a small stream originates. The wetland may at one time have contained a spring. The midden consists of dark, organic‐stained sediments containing animal bone, marine shell, and artifacts. Shell concentrations are evident within the darker midden matrix. These represent individual episodes of shell disposal which eventually began to overlap and coalesce in some parts of the site. The midden’s semi‐circular shape does not appear to be the result of intentional construction, but rather the structured organization of living space around a central “plaza” or common area where several very large pit features and a large clay‐lined hearth were identified. The hearth is located at the bottom of a large, deep pit and may have been used as an earth oven. Other large pits are located around the inner perimeter of the midden. Smaller pit features are more randomly distributed and are located within or immediately below the midden deposits. Several post molds and postholes were identified and mapped leading to the recognition of two possible structures. One possibly oval‐shaped structure measures no more than 2 or 2.5 m in diameter, although it was not entirely exposed. The second structure is larger, with a large open front, and is associated with a hearth. It measures about 7 m across and about 2 m deep. The small sizes of the structures and their relatively insubstantial construction suggests they were inhabited by small family groups. The site contains several very large pits and numerous smaller ones. Some pits were clearly used for refuse disposal although some of the larger ones may have originally been meant to be used for storage. Only one of these is clay‐lined, however, which would have helped to insulate the contents from moisture. Other pits contain no or very little refuse material but are distinguished by their dark organic‐stained sandy fill. Several of these were tested for coprastanol to determine if they were privies, but the results proved negative. The lipid
26
Introduction
analysis did reveal the presence of cholestanol and epicoprostanol in several features, which results from the degradation of terrestrial and marine animal remains, and plant sterols diagnostic of hickory nuts. The lithic concentration near Postl Lake consists of at least two high density areas separated by a moderate to low‐density scatter of artifacts. One high‐density area was excavated by PTA and includes a large pit containing marine shell and lithic tools. Another high‐density area was identified by SEARCH and contains primarily lithic waste flakes. In addition to the PTA feature, five other features were identified. These are mostly refuse pits of various sizes, although the function of two features could not be determined. PTA identified three small hearths or possible hearths in the northeast part of the site and these were associated with lithic waste flakes, small amounts of shell, and a few baked‐clay objects. A total of 16 radiocarbon dates (four obtained by PTA and 12 by SEARCH) identified three temporal occupations while two other possible occupations were identified on the basis of stratigraphy and temporally diagnostic artifacts. The primary occupation was during the Late Archaic period with 14 dates having a maximum two‐sigma range of 4481‐3766 cal BP (2529‐1817 cal BC). Using only the median dates yields a minimum range of 4301‐3971 cal BP (2352‐2002 cal BC), or a little over three centuries. Most of the Late Archaic dates appear to be in good stratigraphic order. The pooled radiocarbon age for the midden is 4078‐3912 cal BP, or 2129‐1963 cal BC. Dated features cluster into three groups. Features 60, 69/104, and 139 range in age between 4150 and 3926 cal BP (2201‐1977 cal BC); Features 44 and 120 range between 4286 and 4000 cal BP (2337‐2051 cal BC); and two shell samples from Feature 4 returned nearly identical ages with a pooled average of 4195 ± 42 (4405‐4149 cal BP, or 2456‐2200 cal BC). This feature is located near Postl Lake and indicates that this area was being used near the beginning of the site’s most intensive occupation during the Late Archaic. Artifacts associated with these dates include a variety of stemmed bifaces, microliths, a large number of baked‐clay objects, and, quartz, quartzite, and sandstone abraders. Along with the radiocarbon dates, these artifacts place the site’s major occupation in the earliest stages of the Elliott’s Point phase of the Late Archaic, as defined by Thomas and Campbell (1991). Two other radiocarbon samples have provided problematic dates. Both samples are from large wood posts. A charcoal sample from Feature 38 returned a radiocarbon age of 250 ± 30 BP. The intercept crosses the calibration curve in several locations ranging from modern (‐1 to 11 cal BP) to between 428 and 376 cal BP (cal AD 1522‐1574), with a 75% probability that the true age is between 428 and 270 cal BP, or cal AD 1522‐1680. If correct, this would date the feature to the Spanish Contact period. The second post, Feature 51, returned a radiocarbon age of 1990 ± 30, or 1997‐1878 cal BP (48 BC‐AD 72). This would situate the feature within the Late Early Woodland period. No artifacts representative of either period have been recovered from Bayou Park with the possible exception of a Fort Walton Incised sherd recovered from a disturbed context, and a possible sherd of Lake Jackson Plain 27
Bayou Park, 8OK898
reported by PTA. Both of these could be associated with the Spanish Contact date, although they fit more comfortably in a pre‐contact, Mississippi‐period context. Woodland‐ period artifacts include sherds of Carrabelle Punctated, Carrabelle Punctated or Ruskin Linear Punctated, and Swift Creek Complicated Stamped pottery. These are diagnostic of the Middle to Late Woodland period and are believed to have been introduced much later than the date from Feature 51. Other ceramic sherds (sand‐tempered plain, indeterminate check‐stamped, indeterminate incised) are neither diagnostic nor especially prevalent. Whatever the age of the ceramic component at Bayou Park, it does not appear to have been very intensive. Except for the two problematic posts, none of the features or midden dates suggests a post‐Archaic occupation. A possible Middle Archaic occupation is suggested by the presence of lithic waste flakes in deeply buried (70+ cmbs) contexts. No temporally diagnostic artifacts were recovered from these depths, although Middle Archaic stemmed bifaces (Marion/Newnan, Putnam, Levy/Pickwick) were recovered from the upper levels of the site and from disturbed spoil. The small numbers of artifacts associated with this component, and their scattered distribution, suggests that it consisted of small, short‐term encampments. Indeed, except for Feature 4, which contained abundant faunal remains, most of the non‐midden portions of Bayou Park were utilized in a similar fashion; that is, as short‐term encampments or specialized activity areas related to the use and maintenance of stone tools. This includes all of the non‐midden areas that are believed to be contemporaneous with the midden based on their stratigraphic position. One of the defining traits of the Elliott’s Point Complex is the presence of baked‐clay objects (also referred to as Elliott’s Point Objects or EPOs). Over 11,000 baked‐clay objects were recovered from Bayou Park, mostly small fragments but over 700 were large enough for detailed analysis. Amorphous forms are the most common although oblate, prolate, tabular, and cylindrical forms also are present. Only a few objects display surface modifications in the form of linear grooves; more common are hand, finger, and fingernail impressions. Excavation of the large clay‐lined hearth recovered a large number of baked‐ clay objects in and around the feature, verifying their use in cooking. Three tabular objects are believed to be parts of griddles used in cooking. One of these had starch grains and phytoliths of bottle gourd, hackberry, possible coontie, and wild grasses embedded in its surface. Analysis of a small sample of baked‐clay objects using X‐ray fluorescence spectrometry suggests that they were made using local clays. Lithic artifacts are dominated by bifacial implements, primarily stemmed bifaces of various types. Use‐wear analysis indicates that bifaces were used for a wide variety of tasks, including cutting, sawing, and scraping, as well as to tip dart points. Bifaces were intensively used, maintained, resharpened, and recycled in an attempt to conserve raw material since no useable tool stone outcrops are located within 100 km of the site. Other chipped stone tools include microliths, a few unifaces, and utilized flakes. Tallahatta Quartzite was the primary raw material used for stone tools. This material outcrops in southern Alabama, 150 km to the north. Cherts from the Marianna and Wrights Creek 28
Introduction
Quarry Clusters also are present but in very low amounts. Quartz and quartzite were obtained from secondary gravel deposits in the Yellow River and its tributaries and sandstone also was obtained from outcrops exposed by alluvial downcutting. The gravels were used for hammerstone/abraders and use‐wear analysis indicates that these implements were likely used in food processing. Most of the sandstone contains iron and one possible grinding stone or receptacle with powdered ochre suggest processing of the sandstone for its iron content in the making of pigments, dyes, or medicines. Faunal remains are dominated by marine resources, particularly oysters, scallops, quahog clams, and a variety of fish species, particularly jacks. Terrestrial resources include deer, rabbit, gray squirrel, raccoon, turkey, crow, ducks, alligator, turtles, gopher tortoise, salamander, and lizards. Eighty‐nine different taxa were identified in the 298,000+ vertebrate and invertebrate specimens that were analyzed. This comprised only a sample of the faunal remains recovered from the site and it is estimated that a minimum of 500,000 remains are available for future research. Plant remains were not numerous but they provide important dietary and seasonality information. Edible plant remains include hickory nut shells, acorn shells, an unidentified seed, and an unidentified fleshy fruit fragment. Pine was the primary wood use for fuel along with hickory and oak. Phytolith and starch grain analysis of a griddle fragment identified bottle gourd, hackberry, possible coontie, as well as phytoliths derived from wild grasses (Chloridoideae, Panicoideae, Bambusoideae, and Pooideae), sedges (Cyperaceae), and from the bark of members of the Annonaceae (custard apple) family. Although the seeds of wild grasses are edible, the grasses may have been present in the surrounding environment or have been used to make fiber. Bottle gourd is a hard‐shelled fruit that can be hollowed out and used as a container and the seeds may be consumed or used to produce oil. The presence of bottle gourd suggests the possibility of limited horticulture. A number of seasonality indicators were identified in the faunal (quahog clam, fish), botanical (hickory nuts and acorns), phytolith (hackberry), and starch grain (coontie) samples. While all seasons of the year are represented by these samples, there is a congruence of all indicators during the early spring through late fall months. This does not rule out the possibility that the site was visited during the winter months, but it appears that occupation during that time of the year was not intensive. The fact that no evidence of substantial structures was found also suggests that occupation occurred during the warmer months of the year. Residential habitation also is indicated by the large pit features and hearths, the presence of probable food processing tools, and the use of local clays to manufacture the large number of baked‐clay objects used in cooking. Residential mobility is indicated by the presence of lithic materials from distant locales and a stone technology that emphasized portability, flexibility, and maintainability. The predominance of Tallahatta Quartzite and small amounts of Suwannee Limestone and Ocala Limestone cherts, indicate that the people who inhabited Bayou Park spent part of the year in the interior where these resources occur. 29
Bayou Park, 8OK898
Only a few pieces of exotic metamorphic stone were recovered from the site, all fragments of ground stone tools, and no steatite, jasper, or exotic cherts have been identified. However, it is likely that the people who occupied Bayou Park interacted with their neighbors along Choctawhatchee Bay. It is noteworthy that no burials have been encountered at Bayou Park. If people were living at the site for several months of the year, it would be expected that some deaths would occur. This suggests that the dead were interred elsewhere, perhaps in a communal cemetery or burial mound. In summary, Bayou Park was occupied primarily during the Elliott’s Point phase of the Late Archaic period sometime between 4481‐3766 cal BP (2529‐1817 cal BC). The site is believed to represent primarily a seasonal (spring‐fall) habitation site used by small kin‐ based groups. Subsistence activities were focused on marine resources, although terrestrial resources also were exploited. Structures were small and of both closed and open construction. Based on the lithic materials from the site, it is hypothesized that people left Bayou Park during the late fall or early winter and traveled north into the interior for the winter months. There they exploited Tallahatta Quartzite and possibly cherts from the Marianna and Wrights Creek Quarry Clusters before moving back to the coast again during the late spring.
CONCLUSION The Bayou Park site has provided a unique opportunity to examine a coastal Late Archaic, Elliott’s Point Complex site in some detail. It has produced important information on the age of the deposits, subsistence, settlement layout, seasonality, and mobility patterns which should assist in a better understanding of Elliott’s Point in the Choctawhatchee Bay region. The following chapters detail these efforts and provide the supporting data for our interpretations of this interesting and significant archaeological site.
30
2
ENVIRONMENTAL AND ARCHAEOLOGICAL CONTEXTS
This chapter provides the reader with environmental and archaeological overviews of northwest Florida and particularly the area surrounding Choctawhatchee Bay. These will provide a context for understanding and interpreting the Bayou Park site (8OK898). The focus of the environmental overview is on those natural resources that would have made the area attractive to Late Archaic‐period populations. To the extent that they are understood, temporal changes in climate, rainfall patterns, sea‐levels and their effect on surface water, plants, and animal life are discussed since significant changes in these would have affected how native peoples adapted their settlement and subsistence strategies to local conditions. This is followed by a review of what is currently known about the pre‐ contact Native American occupation of the region. As extensive and detailed overviews of the environment, paleoenvironment, and prehistory of the Eglin AFB have been presented in previous reports on file with the Cultural Resources Section (see, in particular, Thomas and Campbell 1993, and most recently, Morehead et al. 2011), this chapter will not attempt to repeat these summaries. Instead, we focus on environmental conditions related specifically to the Bayou Park site and its immediate surroundings followed by a discussion of the period of time during which the most intensive occupation of Bayou Park occurred, ca. 4300 B.P. to 3700 BP.
REGIONAL ENVIRONMENTAL SETTING
Physiography The Bayou Park site is located within the Gulf Coastal Lowlands subdivision of the Coastal Plains Province (Brooks 1981; Puri and Vernon 1964). The Coastal Lowlands consists of relatively undissected, nearly level plains which extend from the Gulf coast inland for approximately 16 to 19 km (10 to 12 miles). Maximum elevation in the Coastal Lowlands is typically less than 30 m (100 ft) above mean sea level (amsl). The Coastal Lowlands are believed to have formed as a result of a transgressive sea that eroded into the Citronelle hills, leaving a flat terrace after the sea receded (Schmidt 1978). Subsequent erosion of the highlands and deposition in the lowlands created subtle surface variations (hills and ridges). A portion of the Bayou Park site is situated on one of these low hills, approximately 300 m northwest of Postl Lake (Figure 2.1). The northern portion of the Coastal Plains in northwest Florida is referred to as the Western Highlands. This is a southward sloping plateau that has been dissected by numerous streams resulting in rolling hills and more substantial topographic relief, with elevations reaching as high as 88 m (290 ft) amsl. Many of the streams have steep‐sided valley walls,
Bayou Park, 8OK898
Figure 2.1. Portion of USGS Destin, FL 7.5’ quadrangle map (top) showing the location of 8OK898 within Eglin AFB and a topographic map of the site (bottom) showing the elevated hill system on which much of the Bayou Park site rests.
32
Environmental and Archaeological Contexts
Figure 2.2. Physiographic divisions and subdivisions around 8OK898 and Choctawhatchee Bay (Brooks 1981).
a result of downcutting into the clays that comprise the underlying Citronelle Formation. Brooks (1981) refers to this geomorphic feature as the Eglin Ridge and subdivides it into two sections ‐‐ the Eastern Clay Hills and Western Sandhills (Figure 2.2). The Eastern Clay Hills are more deeply eroded by natural watercourses and exhibit soils with large amounts of clay. The Western Sandhills are much less eroded by surface streams than the eastern portions of the ridge and therefore exhibit thick sand deposits. During the Pliocene and Pleistocene Epochs, expansion and recession of glacial ice masses created episodic fluctuations in sea level, leading to the creation of marine terraces. These terraces are of prime geomorphic importance and are defined more as landscape features than lithologically distinctive stratigraphic or depositional units. The landforms slope gently seaward, often terminating landward by an erosionally produced shoreline scarp. Though debate continues regarding their ages and exact locations, five general terraces have been recognized: Silver Bluff Complex, Pamlico, Penholoway, a high terrace complex of multiple, poorly expressed surfaces, and an upland surface (Johnson and Fredlund 1993). Each of these terraces exhibits some degree of differentiation in the vegetative communities present due to changes in elevations and drainage characteristics. 33
Bayou Park, 8OK898
Bayou Park is situated on the most recent of these terraces, the Silver Bluff. Johnson and Fredlund (1993:37) discuss the various interpretations of the Silver Bluff Complex and conclude that while its base may be primarily Pleistocene in age, its surface has a Holocene component, perhaps due to eolian deposition during the last 5000 years (Johnson and Fredlund 1993:71) Running east/west through Okaloosa County, approximately 10‐15 km north of Bayou Park, is a major marine terrace marked by the Cody Scarp. The Cody Scarp is the northernmost Pleistocene terrace escarpment, a prominent transition between the clay‐rich Miocene soils of the northern Florida panhandle and the Pleistocene sands that typify coastal zones (Platt and Schwartz 1990). Another major geomorphic feature is Santa Rosa Island and its associated lagoons and bays. This large barrier island complex includes river‐mouth swamps and marshes, coastal terraces, the bay, and the barrier bar/island with its active and relict sand dunes, bay‐mouth spits, and submerged shell reefs. Access to the Gulf from Choctawhatchee Bay is through East Pass near Moreno Point. It is believed that this pass narrowed around 1000 BC, reducing the salinity of the bay and altering its shellfish ecology (Goldsmith 1966), an event that is discussed later in greater detail.
Geology, Clay, and Lithic Resources Geologically, the region is covered by surficial sands. The sands are primarily Pleistocene marine deposits that have been reworked and redeposited by aeolian transport and fluvial erosion during the Holocene. According to the Natural Resources Conservation Service (http://websoilsurvey.nrcs.usda.gov/app/WebSoilSurvey.aspx), the moderately well‐ drained Foxworth sand, 0‐5% slopes, is the primary soil type within the boundaries of 8OK898; however, archaeological excavations have identified soil profiles in the southeastern part of the site that resemble Kureb and Resota series (e.g., Mikell et al. 1996:43‐44). The former is excessively drained and the latter moderately well‐drained (http://www.mo15.nrcs.usda.gov/). Both soils are characterized by a loose, sandy composition, with low available water capacity due to very rapid permeability. These surface sands rest on the underlying Citronelle Formation which consists of non‐ marine quartz sands interspersed with gravel and clay lenses (Clark and Schmidt 1982; T. M. Scott et al. 2001). The clay is primarily highly weathered kaolinite containing heavy minerals and mica (Means 2009). Citronelle sediments have been interpreted as representing the deposits of relict rivers that emptied into the Gulf of Mexico during the Pliocene (Clark and Schmidt 1982; Matson 1916; Otvos 2004a). Recent research in Okaloosa and Walton counties (Means 2009) has identified marine fossils in clayey sediments that previously have been assigned to the Citronelle Formation indicating that they have a marine origin. However, since the Citronelle Formation is defined lithologically by an absence of fossils (Clark and Schmidt 1982:33; Marsh 1966:69), it is unclear whether these marine‐derived sediments are part of that formation or are associated with the 34
Environmental and Archaeological Contexts
informally named Miocene Coarse Clastics (Marsh 1966:69). The latter is described as containing sand, clay, gravel, and shells along with the minerals pyrite, mica, and glauconite (Clark and Schmidt 1982:31). These characteristics almost exactly describe the clayey sediments discussed by Means. Regardless of which formation it belongs to, clay exposures are found in cutbanks along the region’s more deeply incised streams and rivers, including the Yellow River which is located along the northern boundary of Eglin AFB (Means 2009:12; Joe Meyer, personal communication, 2012). The Alum Bluff Group is another deposit of sand and clay with traces of heavy minerals, mica, and occasional marine shells (Clark and Schmidt 1982). It grades laterally into the Miocene Coarse Clastics and is known to outcrop in deeply incised streams west of the Yellow River (Means 2009:11). Chert‐bearing limestones of early Miocene, Oligocene, and Eocene age are all subsurface formations with no known exposures in the Coastal Lowlands. In fact, of the numerous chert outcrops recorded across the state of Florida by Upchurch et al. (1982), only two are relatively near the Bayou Park site: the Wright Creek and Marianna Quarry Clusters (QC) (Figure 2.3). Cherts in both of these areas formed through the replacement of limestone by silica which precipitated out of the overlying Hawthorn Formation. The Wright Creek Quarry Cluster consists of several chert outcrops located along the upper Choctawhatchee River and its tributary, Wrights Creek. The Ocala Limestone Formation is the parent rock within which most of the chert from this region has formed. The cherts from this formation tend to have a grainstone fabric and contain numerous foraminifera casts, particularly those of the Orbitoid family (Upchurch et al. 1982:104). Some Suwannee Limestone cherts may also be present. The closest documented chert outcrop is about 70 km northeast of Bayou Park on the south side of Wrights Creek just north of State Road 79. Access to chert in this area could have been accomplished easily by canoe travel across the bay and up the Choctawhatchee River. The Marianna Quarry Cluster is located farther east and includes the upper reaches of the Chipola and Apalachicola Rivers. This group of chert exposures is quite variable with three different chert‐bearing limestone formations present: Tampa/St. Marks, Suwannee, and Ocala. Of these three, the Suwannee Limestone cherts seem to be most diagnostic and these cherts are present at the well‐known Two Egg Quarry located about 145 km northeast of Bayou Park. Chert from this quarry is described as “gray, pink, and lavender, with gray‐ colored material predominating” (Sharon and Watson 1971:77). Upchurch et al. (1982:106) also describe cherts from his site as light‐colored with low iron content and few observable fossils. Light‐colored cherts found on sites located on Eglin AFB and surrounding area often are referred to as “Two Egg chert.” Although there is no data to indicate that these were derived from the Two Egg quarry, they do fit the description of Suwannee Limestone cherts from the Marianna Quarry Cluster generally. 35
Bayou Park, 8OK898
Figure 2.3. Map of geological formations in the southeastern U.S. that contain knappable toolstone. Modified from Price (2008).
Another important source of toolstone for prehistoric people in the Florida panhandle is the Tallahatta Formation (Haywick and Carr 2004; Price 2008). This formation outcrops in a band that extends from the Chattahoochee River in southwest Georgia west through southern Alabama and into eastern Mississippi. It contains localized deposits of silicified sandstone, referred to in the archaeological literature as Tallahatta Quartzite, which are exposed in stream valleys (Lloyd et al. 1983). The material is typically a dull, opaque tan in color with numerous quartz grains which provide it with a glittery appearance. According to Lloyd et al. (1983:127), the outcrops are thickest in western Alabama in the counties of Clark, Washington, and Choctaw. At a minimum, this would place these source areas 120 to 130 km from Bayou Park. Outcrops are present along the lower Chattahoochee in southeast Alabama and southwest Georgia (Nancy White, personal communication 2012) and could have been accessed by coastal peoples via the Apalachicola River. Several sites along the Apalachicola River that contain artifacts made of Tallahatta Quartzite have been documented (J. S. Austin 2003). Finally, the Citronelle Formation contains chert and quartz gravels that could have been used for stone tool production, although extensive evidence for the use of this material is lacking at sites around Choctawhatchee Bay. Means (2009:39) notes that no chert gravels were identified in any of his sample locations, although quartzite gravels, some up to 5 cm in diameter, are present in some locations (2009:41). 36
Environmental and Archaeological Contexts
Hydrology Three major hydrologic basins are present within the Eglin AFB: Choctawhatchee Bay, Yellow River Basin, and Pensacola Bay (Science Application International Corporation [SAIC] 2010:4‐6). The Bayou Park site is influenced primarily by the Choctawhatchee Bay basin, which in turn is influenced by the western panhandle aquifer, rainfall patterns, and the watershed of the Choctawhatchee River.
Aquifer and Rainfall The upper aquifer of the Florida panhandle is dominated by a surficial sand and gravel aquifer. The aquifer is divided into three zones with alternating permeabilities (Miller 1997:75). The uppermost layer, referred to as the surficial zone, is highly permeable and is recharged by precipitation. Most of the water in this zone moves laterally along the less permeable middle zone and is discharged as base flow in streams or as steepheads. Where they are well‐developed and unaffected by logging and subsequent erosion, steephead springs produce a significant amount of potable water and would have provided a valuable resource to prehistoric populations (Fredlund and Johnson 1993). Rainfall also percolates through the middle zone and recharges the third, or lowest, zone. Water contained within this lower permeable zone is under artesian pressure due to the overlying confining layer of marl and clay, of which the Citronelle Formation is a part. This lower zone is the main water producing zone for the region and it is tapped by numerous wells. Natural discharge occurs where streams have dissected the landscape and intercepted the lower zone; these streams channel water to the Bay and Gulf. The upper sand and gravel aquifer is separated from the Floridan Aquifer by another confining unit of clay, the Pensacola Clay (SAIC 2010:4‐12). Unlike to the east of Choctawhatchee Bay in the Marianna Lowlands, the Floridan Aquifer is buried beneath several hundred feet of sand and clay and is not a major source of water for coastal panhandle residents. Along the northern reaches of the Choctawhatchee River, however, the Floridan Aquifer is at or near the surface and is frequently exposed in the river’s channel as it nears the Alabama border (Barrios 2005). The saturated thickness of the sand and gravel aquifer, based on current water conditions, varies between 15 and 38 meters. Variation in the thickness is heavily dependent on seasonal rains, with saturation thinning during periods of extend drought or thickening during increased precipitation (DeFosset 2004:1). The climate of coastal Okaloosa County is characterized as subtropical with humid, warm summers and mild winters. Yearly rainfall averages 157.5 cm, with a monthly average of 12.9 cm. Much of the annual rainfall (44%) occurs during the months of July through September, with most falling during the months of July and September (Overing et al. 1995:1; SAIC 2010:4‐1). The months of October through February show the lowest trends in 37
Bayou Park, 8OK898
rainfall. Extended dry periods can occur during any season but are most frequent during the spring and fall. In general, dry periods during the fall tend to be longer in duration, but the shorter spring dry periods often are more severe due to rising average temperatures and a greater demand for moisture by ecosystems (Overing et al. 1995:2).
Choctawhatchee Bay Choctawhatchee Bay is a large linear estuary stretching east to west for about 50 km across Walton and Okaloosa Counties. The bay ranges in width from 2 to 10 km and in depth from 3 to 13 meters. The primary source of freshwater for the bay is the Choctawhatchee River which spills into the bay via a delta at its eastern end. The Choctawhatchee River contributes the majority of the bay’s 13,856 sq km watershed, which expands northwest through Florida and into southern Alabama (Ruth and Handley 2007:143). The Choctawhatchee River is fed by 13 identified springs, both the typical fissure‐type vents and those described as seep springs, where ground water percolates upwards into diffuse pools (Barrios 2005:6). The annual average discharge rate of freshwater into the bay from the river is approximately 243 m3/second (Ruth and Handley 2007:143). Other sources of fresh water include Turkey Creek, Rocky Branch, Mullet Creek, Trout Creek, Lafayette Creek, Ramsey Branch, Black Creek, Mitchell River, among other smaller streams. The principle marine access and only direct route to the Gulf of Mexico is East Pass, located between Destin and Okaloosa Island, approximately 9 km south‐southeast of Bayou Park. According to Goldsmith (1966), this pass was open prior to about 1000 BC but became restricted as a result of westward longshore drift that formed a barrier spit known today as Moreno Point. An 1886 coastal navigational chart shows that only a narrow pass connected the two water bodies, but by 1922 this had become little more than an intermittent stream (Office of Coast Survey 1886, 1922). When in 1929 major flooding of the Choctawhatchee River caused bay waters to rise precipitously, local residents reportedly dug a trench along this stream to enable bay waters to escape into the Gulf (Ruth and Handley 2007:145). The US Army Corps of Engineers widened the pass in 1930 and has maintained it since. Access to the Gulf also is provided by Santa Rosa Sound just northwest of East Pass, and by an intracoastal waterway at the Bay’s east end, just south of the Choctawhatchee River. According to Hoyer et al. (2013:455), the opening of the pass increased salinity in the bay resulting in a loss of freshwater marshes and sea grass beds.
Boggy Bayou Situated north of Bayou Park, between the cities of Valparaiso and Niceville, Boggy Bayou opens into the northwest side of Choctawhatchee Bay and extends to the northwest for about 5 km. The elongated bayou varies in width from approximately 0.38 km at its narrowest and 0.95 km at its mouth with depths ranging 4 to 6.7 meters (Campbell et al. 1989:1). The bayou is fed primarily by Tom’s Creek and Turkey Creek. The overall salinity of Boggy Bayou ranges from 10 to 15 ‰ at the surface to 20 to 25 ‰ at the bottom (Butts 1997:4). Near the mouth of Boggy Creek and just north of Bayou Park is Weekley Bayou which is fed by a small stream. 38
Environmental and Archaeological Contexts
Postl Lake The south and west boundaries of the Bayou Park site are formed by drainages which empty into Postl Lake. The lake is named for Charles Postl (Postil), formerly of Chicago, who owned and operated a health resort in the 1920s and 1930s (Baker Block Museum n.d.). The narrow L‐ shaped lake is actually a small lagoon extending northeast and then north, paralleling Choctawhatchee Bay. It is connected to the Bay at its north end by a narrow drainage and on the east by a shallow pass. It is unclear if this lagoon was present at the time of prehistoric occupation of Bayou Park Figure 2.4. Portion of 1935 USGS Villa Tasso 15’ or if there was ever a time when it quadrangle map showing Postl Lake. was not connected to the bay. Early maps of the area show no lake or lagoon in this location (Campbell, Morehead, et al. 2009:Figures 14‐16; General Land Office 1851; Office of Coast Survey 1886;). The first map that shows Postl Lake is a 1935 USGS quadrangle map (Figure 2.4) which indicates that the lake was landlocked at that time. A 1955 USDA aerial photograph (Figure 2.5) shows a small drainage at the lake’s north end and what appears to be a gap in the lake’s eastern shoreline connecting it to the bay. When this gap began to intercept the lake is unknown. The USDA aerial also shows clearly the prograding nature of the shoreline east of Postl Lake toward the mouth of Boggy Bayou. The lake’s scalloped eastern shoreline also is visible. Scalloped shorelines are features seen on the lagoonal sides of barrier islands and are usually formed by washover fans during major storm events (Liu and Fearn 2000). These observations suggest that Postl Lake may have formed as a result of a prograding spit which eventually closed off the lagoon. Archaeological site 8OK72 on the south side of the lake contained fiber‐tempered sherds as well as later artifacts (Lazarus 1964) suggesting that a lake or a marsh existed at least as early as 1000 BC.
Flora The portion of the western panhandle that contains Eglin AFB and the Bayou Park site is contained within the Sandhill High Pine region (Myers 1990:Figure 6.14). The sandhill high pine region extends from the Suwannee River westward into Alabama and Mississippi. Myers (1990:177‐186) and Harper (1914:233‐238) provide general descriptions of the vegetation which, as the name suggests, is characterized by forests of longleaf pine (Pinus palustris). Harper’s description was made prior to modern development, although the area had already experienced intensive timber and naval stores harvesting during the nineteenth century (Harper 1914:239‐240). This early commercial exploitation plus modern fire 39
Bayou Park, 8OK898
Figure 2.5. USDA aerial photograph showing Postl Lake in 1955.
suppression has led to the general replacement of longleaf pine by xerophytic oak species and the development of xeric hammocks. The Eglin AFB Integrated Natural Resources Management Plan (INRMP) divides the base into four broad ecological associations: sandhills, flatwoods, barrier islands, and wetlands (SAIC 2010). The sandhills association is the most extensive, covering over 360,000 acres. It is characterized by an open, savanna‐like structure with a canopy of longleaf pine, a sparse midstory of oaks and other hardwoods, and a groundcover comprised mainly of grasses, forbs, and low shrubs. The pine sandhills contain a high diversity of fire‐adapted species. Other important flora include a variety of xeric‐adapted deciduous oaks, such as turkey oak (Quercus laevis), live oak (Quercus virginiana), and blackjack oak (Quercus marilandica), and herbaceous ground cover, of which wiregrass (Aristida stricta) and little bluestem (Schizachyrium scoparium) are the most common. Associated with pine sandhills are xeric hammock, scrub, upland pine forest, and slope forests. Xeric hammocks are typically characterized as scrubby, low canopy forests with little understory other than saw palmetto (Serenoa repens). Typical botanical taxa include live oak, sand live oak (Quercus geminata), Chapman’s oak (Quercus chapmanii), turkey oak, southern magnolia (Magnolia grandiflora), sparkleberry (Vaccinium arboreum), beautyberry (Callicarpa americana), and yaupon (Ilex vomitoria) (Florida Natural Areas Inventory [FNAI] and Florida Department of Natural Resources [FDNR] 1990:6‐7). Scrub communities are characterized by sand pine (Pinus clausa), sand live oak, scrub live oak (Quercus inopina), myrtle oak (Quercus myrtifolia),
40
Environmental and Archaeological Contexts
Chapman’s oak, saw palmetto, and rosemary (Ceratiola ericoides) (FNAI and FDNR 1990:5‐ 6). Slope forests are unique mesic settings that result from the formation of steepheads, which are steep and narrow valleys formed by the downcutting action of flowing springs or seeps. Steephead springs typically produce well‐developed, closed canopy hardwood hammocks on steep, sheltered slopes. These natural communities are common in the uplands of the Florida panhandle (Means 1991), and Thomas and Campbell (1993) have observed that steepheads are present over a large area of Eglin AFB in Okaloosa County. In addition to potable water, the combination of a cool, moist microclimate and the densely shaded slopes of a steephead results in conditions conducive to the growth of several species more typical of the Piedmont and Southern Appalachian Mountains (FNAI and FDNR 1990:15). Typical slope forest species include black walnut (Juglans nigra), basswood (Tilia americana), strawberry bush (Euonymus americanus), Florida yam (Dioscorea floridana), Solomon’s seal (Polygonatum biflorum), leatherwood (Dirca palustris), American beech (Fagus grandifolia), white oak (Quercus alba), laurel oak (Quercus hemisphaerica), mockernut hickory (Carya tomentosa), sarsaparilla vine (Smilax pumila), witchhazel (Hamamelis virginiana), horse sugar (Symplocos tinctoria), and Florida yew (Taxus floridana), among many others (FNAI and FDNR 1990:14‐15). Pine flatwoods occur on flat, moderately well‐drained sandy soils that are often underlain by a hardpan. The canopy consists of slash pine (Pinus elliottii) and longleaf pine, with an understory that varies from dense shrubs, particularly saw palmetto, to open grasses and herbs. Soil moisture and fire history are the main factors that influence the types of vegetation that occur in pine flatwoods. Interspersed within the pine flatwoods community along the coast are maritime hammocks, one of which is located within the southern boundary of 8OK898. These are typically narrow bands of hardwood forest characterized by live oak, cabbage palm (Sabal palmetto), redbay (Persea borbonia), southern magnolia, American holly (Ilex opaca), sea grape (Coccoloba uvifera), saw palmetto, and beautyberry (FNAI and FDNR 1990:11‐12). The natural communities associated with the barrier island complex include primary and secondary dunes, interdune swales, maritime forests, and sand pine scrub. The vegetation that occupies these communities is influenced by coastal (maritime) processes such as erosion, deposition, salt spray, and storms. Wetland types on the Eglin AFB are varied and include baygall, seepage slope, dry prairie, flatwood lake, floodplain forest, floodplain swamp, bottomland forest, wet prairie, hydric hammock, blackwater stream, alluvial stream, spring run stream, steepheads, marsh lake, slough, dome swamp, strand swamp, basin marsh, depression marsh, floodplain marsh, sandhill upland lake, bog, freshwater tidal swamp, and salt marsh (SAIC 2010). Artesian spring‐fed streams are common along the upper reaches of the Choctawhatchee River 41
Bayou Park, 8OK898
(Barrios 2005), but most of the streams near 8OK898 are seepage streams or blackwater streams. Steepheads are clear to lightly‐colored, relatively short, shallow, and narrow water courses that originate from shallow groundwaters that have percolated through deep, sandy uplands. Most are perennial and flow from north to south towards the coast. Most of the natural ponds and non‐forested wetlands are usually relatively small. Some hold water permanently, while others are seasonal. Some contain herbaceous or woody vegetation (SAIC 2010:4‐6, 4‐9). Figure 2.6 shows the distribution of wetlands on Eglin AFB in relation to 8OK898. Marine and marsh environments associated with Choctawhatchee Bay would have been a critical source of subsistence and other resources for the prehistoric inhabitants of Bayou Park. Saltwater marsh areas along the bay shores exhibit a diverse grouping of botanical taxa. These taxa include black needlerush (Juncus gerardii), smooth cordgrass (Spartina alterniflora), saltgrass (Distichlis spicata), marsh elder (Iva annua), and cattail (Typha).
Fauna The sandhill high pine, pine flatwoods, and Choctawhatchee Bay provided a wide range of animal species for prehistoric fishers and hunters. Terrestrial species that would have been commonly available to native populations in both the sandhill and the pine flatwood environments include white‐tailed deer (Odocoileus virginiana), Florida black bear (Ursus americanus floridanus), bobcat (Lynx rufus), gray fox (Urocyon cinereoargenteus), raccoon (Procyon lotor), opossum (Didelphis virginiana), box turtle (Terrapene carolinia), and eastern diamondback rattlesnake (Crotalus adamanteus) (SAIC 2010:Table 5‐2). The bayous branching out of Choctawhatchee Bay and boarding Bayou Park would have also provided access to alligator (Alligator mississippiensis) and numerous species of aquatic turtles, such as alligator snapping turtle (Chelydra serpentine), Florida soft shell (Apalone ferox), pond sliders (Trachemys scripta), coastal plain cooter (Pseudemys floridana), and mud and musk turtles (Kinosternidae). The presence of highly productive seagrass beds and variable salinities due to the intermixing of freshwater from the Choctawhatchee River and Gulf tidal waters creates a diverse range of habitats for both invertebrate and vertebrate resources. The benthic habitat of the bay is dominated by low relief sand and mud flats interspersed with oyster and seagrass beds (Fox et al. 2002:21). Figure 2.7 reproduces a map of seagrass beds in Choctawhatchee Bay ca. 1972 presented in Ruth and Handley (2007). It shows that most seagrass beds are located in the western bay, near East Pass. Seagrass beds are noticeably absent around the mouth of Boggy Bayou, a situation that was also noted by Butts (1997) in his survey. These researchers indicated that this is likely a result of high levels of heavy metals probably from a nearby boat yard (Butts 1997:21). Common species that would likely have been available in the bay include eastern oysters (Crassostrea virginica), shrimp (Penaeus spp.), blue crab (Callinectes sapidus), largemouth bass (Micropterus salmoides), spotted seatrout (Cynoscion nebulosus), Gulf menhaden 42
Figure 2.6 Map of Eglin AFB showing the distribution of wetlands in relation to Bayou Park (8OK898). SOURCE: Science Application International Corporation 2010:Figure 4‐4.
Environmental and Archaeological Contexts
43
Bayou Park, 8OK898
Figure 2.7. Seagrass beds in Choctawhatchee bay ca. 1972. Reproduced from Ruth and Handley (2007).
(Brevoortia patronus), red drum (Sciaenops ocellatus), Gulf flounder (Paralichthys albigutta), striped mullet (Mugil cephalus), white mullet (Mugil curema), and Gulf sturgeon (Acipenser oxyrinchus) (Ruth and Handly 2007:143).
PALEOENVIRONMENTAL CONDITIONS
Late Pleistocene‐Early Holocene From 18,000 to 15,000 BP, immediately following the Pleistocene glacial maximum, Florida’s climate and vegetation were influenced by the effects of continental glaciers located farther north and by the release of meltwater and ice rafts into the oceans as worldwide climate began to moderate (Alley and Clark 1999; Clark et al. 1999). Relatively cool conditions are believed to have prevailed until about 13,000 BP (Watts and Hansen 1994). Gates (1976) has suggested that the temperature may have been as much as 8° to 10° C cooler. Pollen evidence from Florida lakes combined with data from deep sea cores (Grimm et al. 1993) suggest variable wet and dry conditions during this time. At about 13,000 years ago, the climate cooled and extreme aridity may have occurred as a result of the release of massive amounts of glacial meltwater into the Atlantic Ocean. Dunbar (2002:160‐161) reports that water levels in the Aucilla River were extremely low between 13,000 and 12,700 BP. He also notes a retreat of spruce in north Florida during this time, an indication of decreased moisture. Vegetation in the Southeast lost much of its diversity. Dense mixed hardwood forests expanded, replacing the northern arboreal taxa which had proliferated during the preceding period (Delcourt and Delcourt 1981). Water levels continued to be depressed in both north and south Florida, due to reduced rainfall and low sea levels, until about 11,700 BP when wetter conditions returned in north Florida 44
Environmental and Archaeological Contexts
(Dunbar 2002). In south Florida, environmental conditions may have been slightly different. The region apparently continued to experience a dry climate and a depressed water table during this time (Grimm et al. 1993; Watts and Hansen 1988, 1994), although some surface water apparently accumulated in shallow basins (Dunbar 2002:165, 173‐174). Using pollen data from Lake Tulane in Highlands County, Grimm et al. (1993:199) show oscillations in wet and dry conditions, but the magnitude of the oscillations is not as great as during the preceding 10,000‐year period. From 10,000 BP to 8500 BP, relatively dry conditions are believe to have been present in central and south Florida (Watts et al. 1996; Watts and Hansen 1988:311‐313). At Lake Tulane, large amounts of charcoal are present in sediments dating between about 11,000‐7000 BP indicating a prevalence of fires, which suggests that an abundance of dry fuel was present (Watts and Hansen 1988:311). The inference is that precipitation was minimal during this time. This is supported by the absence of organic sedimentation at other lakes in southern Georgia and northern and central Florida prior to 8500 BP (Watts 1969, 1971, 1975; Watts and Hansen 1988:311‐313). In effect these lakes were dry basins. This pattern is repeated at Little Salt Springs and Warm Mineral Springs in Sarasota County. Clausen et al. (1979:610‐611) report no aquatic or marsh plants in pollen dated to 9920 BP at Little Salt Springs and dry conditions also are indicated at Warm Mineral Springs between 9870‐8520 BP (Clausen et al. 1975:197). In north Florida, Dunbar (2002:175‐176) indicates that dry conditions prevailed there between 10,000 and 9000 B.P., but that water levels gradually began to rise shortly thereafter, reaching near‐modern levels by 8500‐8000 B.P. (Watts 1969, 1971, 1983; Watts and Stuiver 1980). Fredlund and Johnson (1993) conducted fossil pollen analyses at four sites on Eglin AFB, providing local paleoenvironmental data and allowing for the reconstruction of the Holocene history of the pine‐oak forests of the region around 8OK898. The evidence suggests that, as with patterns observed regionally, the local climate shifted from one of less annual rainfall to one of more mesic conditions with a more seasonally variable moisture regime around 8400 BP. The changes in pollen percentages and accumulation rates which were observed for tree and shrub taxa document a 1200‐year period of vegetative readjustment following the onset of these climatic changes. Prior to this period an open oak‐pine forest containing thickets of xeric shrubs were present. With the onset of a more mesic climate, pines began to increase in abundance. Fires associated with the new climatic regime selected for fire‐tolerant longleaf pine seedlings rather than those of other tree species. These selective processes continued to shift the proportions of arboreal taxa until about 7200 BP, when the longleaf pine forests reached the state of dynamic equilibrium that was observed historically within the region (Fredlund and Johnson 1993:120‐121).
Sea‐Level Rise Associated with these climatic shifts, was a major global rise in sea level as glaciers melted (summarized for Florida by Balsillie and Donoghue 2004; Dorsey 1997). At 18,000 BP sea‐ 45
Bayou Park, 8OK898
level stood at ‐120 meters. The subsequent rise was slow while glacial conditions prevailed at high latitudes but became very rapid during the latest Pleistocene and earliest Holocene, rising nearly 100 m between 15,000 and 8000 BP (Dorsey 1997:Figure 5). The major changes in Pleistocene versus Holocene sea levels affected Florida in three major ways. First, because of the very broad, gently sloping continental shelf on the Gulf Coast, higher sea levels greatly reduced the land area of the Florida Platform. Second, given the region’s porous limestone bedrock, the higher sea levels of post‐glacial times resulted in much higher groundwater levels in the artesian aquifer. This, coupled with increasing rainfall and warmer climate during the Holocene, had a dramatic effect on water resources for animals and humans. Third, more abundant water and a warmer, wetter climate combined to create more modern native plant communities. As eustatic sea level rise slowed and began to fluctuate after about 8000 B.P., stream systems cut multiple steepheads and then stabilized; their valley bottoms filled with primarily pine detritus from the surrounding forest vegetation. Santa Rosa Island formed during this time as well (Johnson and Fredlund 1993).
Mid‐to‐Late Holocene Modern types of native plant communities, dominated by coniferous forests, mixed coniferous‐deciduous forest, deciduous forest, and various types of wetland vegetation, began to develop around 8500‐8000 BP (Fredlund and Johnson 1993; Watts et al. 1996; Watts and Hansen 1988). These conditions have been fairly stable since that time, although periodic, short‐term fluctuations in climate, precipitation, and sea level have occurred as a result of glacial expansion and contraction. At the local scale, these climatic fluctuations resulted in conditions that were both wetter and drier than today. For example, Gleason et al. (1974:311) report evidence of ancient fires in a peat bog in Marion County, and from this infer four separate periods of severe drought between 4030 and 2900 B.P. Fredlund and Johnson (1993:118) report that cores from Eglin AFB display perturbations in the ratios of pine to oak pollen after 7200 BP and they suggest that these are related to short‐term climatic fluctuations. Sea level also was higher during this time, although estimates of how high vary among researchers. Fairbridge (1974) has estimated that it was 3‐4 m higher than present. Using data from nearby St. Vincent Island, Stapor and Tanner (1977) suggest a mid‐Holocene high of 1.5 m above present levels (see also Stapor et al. 1991; Tanner 1991). Scholl et al. (1969:Figure 2) offer a more conservative estimate of 4 m below present levels while Faught and Donoghue (1994) placed sea level at +2.45 m at 5500 BP. By 4000 BP the climate, water levels, and plant communities of Florida attained essentially modern conditions, although multidecadal‐ and century‐scale variability did occur (Mayewski et al. 2004; Poore 2008; Soto 2005). Of particular interest for this study is the evidence for a cool, dry event between 4200 to 3800 BP followed by a period of extreme aridity between 3500 and 2500 BP. These climatic events are found in the global climate 46
Environmental and Archaeological Contexts
record at temporally equivalent periods (Mayewski et al. 2004), but their local effects were variable and, for Florida, not well understood (Poore 2008). The rate of sea‐level rise also slowed and many of the barrier islands around the Gulf began to form near their present position (Williams et al. 1999). The slowing of sea‐level rise allowed progradation of several river deltas around the Gulf of Mexico that currently support coastal swamp forest. For example, swamp and bottomland forest near the Apalachicola River occupy a delta that appears to have expanded since 4000 BP (Donoghue and White 1994). Several studies have provided evidence of higher and lower sea‐level stands during the past 3000 years (e.g., Balsillie and Donoghue 2004; Fairbridge 1974; Stapor et al. 1991; Tanner 1991, 1992; Walker et al. 1994, 1995). Dating of shells in beach ridges on St. Vincent Island and on barrier islands in southwest Florida indicate that the ridges were deposited during discrete periods which are interpreted as resulting from rises in sea level at ca. 2000 B.P., 1100 B.P., and 150 B.P. (Stapor et al. 1991; Tanner 1991, 1992). These periods of higher sea level correspond with periods of warmer temperatures, the Roman Warm Period, the Medieval Warm Period, and modern climatic conditions, respectively. Around Apalachicola Bay, evidence of human habitation alternates with strata of estuarine muds, suggesting fluctuations in sea‐level with associated movements in the position of the habitable shoreline (Donoghue and White 1994). Johnson and Fredlund (1993:52) note the similarities in sea level data from St. Vincent Island with data from Eglin AFB, with the latter including the proximity of shell middens to the present shoreline, submergence of middens below present sea level, and the vertical position of paleosols and tree stumps relative to modern sea level. Using these combined date, Johnson and Fredlund present a model of Holocene sea level fluctuation for Eglin AFB that is reproduced here as Figure 2.8. According to their reconstruction, sea level during the Late Archaic period, when Bayou Park was most intensively occupied, would have been approximately two meters above present levels but falling at approximately 4000 BP, reaching a low of ‐2 meters at about 3000 BP. Sea level then began another slow rise to about +1 meter at around 1000 BP before falling again to – 1 m during the Fort Walton period.
Moreno Point Presently, Moreno Point is the end of a barrier spit located in Destin on East Pass, almost directly across Choctawhatchee Bay from 8OK898. It has been hypothesized that the formation of Moreno Point had a significant environmental impact on local coastal areas, specifically the salinity of the Bay and the effect this had on local shellfish populations (Goldsmith 1966; Thomas and Campbell 1991:114‐115, 1993:539‐540). Based on the results of his sedimentary study of Choctawhatchee Bay, Goldsmith (1966) hypothesized that sometime between 7000 and 3000 BP, Moreno Point began to form as the result of westward longshore drift, eventually becoming a barrier spit. By about 3000 BP, the spit had reduced the opening of Choctawhatchee Bay to the Gulf of Mexico sufficiently enough to restrict water exchange between the two bodies of water, resulting in natural 47
Bayou Park, 8OK898
Figure 2.8. Reconstructed Holocene sea level fluctuations for the Eglin Air Force Base/Choctawhatchee Bay region. Reproduced from Johnson and Fredlund (1993:Figure 25).
desalination of the bay. The bay’s reduced salinity ultimately resulted in environmental conditions unsuitable for halophilic (favoring high salinity) species of shellfish while also providing niches for species that are more freshwater tolerant. The slow accretion of Moreno Point and related morphological events are cited as having influenced the prehistoric populations that occupied the shores of Choctawhatchee Bay throughout the Late Holocene (Thomas and Campbell 1991, 1993). The lower salinity in the bay resulting from the formation of Moreno Point is suggested as the factor responsible for the differences in shellfish composition between Elliott’s Point (Late Archaic) and Deptford (Early Woodland) middens. The presence of bay scallop (Argopecten irradians) and quahog (Mercenaria spp.) in earlier Elliott’s Point middens, two species that favor high salinity habitats, is seen as evidence that Choctawhatchee Bay was open to the Gulf during this time period. The subsequent absence of scallop and quahog in the middens of later Deptford populations is offered as support of the hypothesis that by 2500 BP the salinity of the bay had decreased substantially as a result of the accretion of Moreno Point. It appears that the slow accretion of the barrier spit at Moreno point continued through the Middle Woodland period, reducing the salinity levels to a point that even eastern oyster (Crassostrea virginica), a species that can tolerate varying salinity levels, could no longer survive. Rangia cuneata, a species of marsh clam that thrives in waters with very low salinity, is the dominant species represented in Middle Woodland (Santa Rosa/Swift Creek) 48
Environmental and Archaeological Contexts
middens throughout the Choctawhatchee Bay system (Thomas and Campbell 1993:564‐ 565). A return to oyster utilization is recorded archaeologically for Weeden Island (Late Woodland) and Fort Walton (Mississippian) populations, and Thomas and Campbell (1993:565) posit that sometime between the Middle and Late Woodland periods, a significant event, such as a major hurricane, reopened the pass, increasing bay water salinity to levels suitable for oyster.
PREHISTORIC OVERVIEW The prehistory of Choctawhatchee Bay and Eglin AFB has been summarized by Thomas and Campbell (1993:489‐637) with periodic updates presented in various reports since that time. Table 2.1 is their most recent update of the region’s cultural chronology and is reproduced from Morehead et al. (2011:Figure 4). As the Bayou Park site was occupied primarily during the Late Archaic period, the following sections provide a review of Late Archaic/Elliott’s Point developments followed by a brief summary post‐Archaic developments in the panhandle. Before proceeding, however, we must first note our use of the term Late Archaic in place of Thomas and Campbell’s Gulf Formational when referring to the period of time from about 5000 BP to 2500 BP (3000 BC to 500 BC). As discussed in detail by Thomas and Campbell (1993:518‐522), the Gulf Formational concept was first put forward by Walthall and Jenkins (1976) to refer to the period of time when fiber‐tempered pottery first began to appear in the Southeastern US. While it is formally used in Alabama and the western Florida panhandle, it has not seen consistent usage elsewhere nor is it referenced in the most recent regional syntheses (e.g., Bense 1994; Milanich 1994; Sassaman 2010). While arguing in favor of its usefulness at the local level, Thomas and Campbell (1993:521) acknowledge that the absence of reliable dates for the appearance of fiber‐tempered pottery in the panhandle may have contributed to the reluctance of some archaeologist to use it in their temporal reconstructions. As they note, there are no dates from the Choctawhatchee Bay region which would place the appearance of fiber‐tempered pottery any earlier than 3135 BP (1675‐1025 cal BC), and they conclude that fiber‐tempered pottery was a late addition to the already existing Late Archaic/Elliott’s Point material culture (Campbell et al. 2004:132; Thomas and Campbell 1993:528, 538, 541). On the other hand, White (2003) reports fiber‐tempered pottery as early as 3970 radiocarbon years (rcy) BP in the Apalachicola River region, or 2900‐1980 cal BC. Further, Thomas and Campbell (1993:540) admit that the characteristics of Elliott’s Point “do not fit neatly within the Gulf Formational framework.” While we do not intend to enter the debate on whether the Gulf Formational concept is applicable or useful, we have decided to use the more general Late Archaic designation when referring to the period of time that encompassed the Elliott’s Point Complex. This decision is based on its more uniform usage throughout Florida and the SEUS. Its use also 49
Bayou Park, 8OK898
Table 2.1. Native American culture sequence in the Choctawhatchee Bay region (modified from Morehead et al. 2011:Figure 4). STAGE
CALENDAR YEARS
PERIOD
CULTURE
Historic
Mississippian
Woodland
Gulf Formational
Archaic
Lithic
50
AD 1800— AD 1700— AD 1600— AD 1500— AD 1400— AD 1300— AD 1200— AD 1100— AD 1000— AD 900— AD 800— AD 700— AD 600— AD 500— AD 400— AD 300— AD 200— AD 100— — 100 BC — 200 BC — 300 BC — 400 BC — 500 BC — 600 BC — 700 BC — 800 BC — 900 BC — 1000 BC — 2000 BC — 3000 BC — 4000 BC — 5000 BC — 6000 BC — 7000 BC — 8000 BC — 9000 BC — 10000 BC — 11000 BC —
Historic Late Mississippian Fort Walton/ Pensacola
Middle Mississippian Early Mississippian Late Woodland
Weeden Island
Middle Woodland
Santa Rosa/ Swift Creek
Early Woodland
Deptford
Gulf Formational
Elliott’s Point/ Norwood
Early to Middle Archaic
Paleoindian/Early Archaic
PHASE/COMPLEX
Four Mile Point Indian Bayou Horseshoe Bayou Lassiter Okaloosa Alligator Lake Elliott’s Point
Environmental and Archaeological Contexts
avoids the issue of the timing of the appearance of fiber‐tempered pottery discussed above as a factor in the definition of the Gulf Formational. If the primary reason for using the Gulf Formational term is to highlight the relationships between Late Archaic societies that inhabited the Florida coastal panhandle with contemporaneous Late Archaic groups, then the Elliott’s Point designation satisfactorily achieves this goal. Indeed, Thomas and Campbell (1993:528) have stated that since fiber‐tempered ceramics in the region tend to be found in settings that also contain Elliott’s Point sites, “we are considering all Gulf Formational sites to be possibly related to Elliott’s Point” (see also Morehead et al. 2011:20).
The Elliott’s Point Complex The peoples who lived in the Southeastern United States during the Late Archaic period (approximately 5000‐3000 BP, or 3000‐1000 BC) exhibited a variety of cultural characteristics and organizational systems (Anderson 1996). Traits such as the production of fiber‐tempered ceramics, degree of residential mobility, investment in fixed territories, and degree of organizational complexity were manifested in different ways among different groups regardless of intergroup distance (Anderson 1996:157, 166‐167). Many Late Archaic cultures participated in long‐distance interaction within the Poverty Point Exchange Network, a vast yet discontinuous system of exchange centered on the Poverty Point site in northeastern Louisiana (Figure 2.9) (Gibson 2000). The cultural diversity that existed within the Late Archaic Southeast largely resulted from differences in physiography, resource structure, and degree of interaction between neighboring social groups (Anderson 1996:176). The Elliott’s Point Complex is a local manifestation of Late Archaic cultural traditions that was centered around Choctawhatchee Bay. It was first defined by Lazarus (1958) as a geographically bounded collective of sites exhibiting a combination of Late Archaic assemblage characteristics, including an absence of pottery, the presence of stemmed projectile points, and more importantly, the presence of baked‐clay objects and microlithic “drills” and “perforators.” Lazarus (1958:26) noted the similarities between Elliott’s Point and Poverty Point, where microliths and baked‐clay objects both were utilized, although he also noted differences between the two in terms of their baked‐clay‐object assemblages. Lazarus named the local complex Elliott’s Point after the location of the sites where he first identified these traits. Since its initial taxonomic designation, the Elliott’s Point Complex has come to include a developmental culture sequence based on artifact assemblage characteristics and presumed participation within the Poverty Point Exchange Network, evinced by the material correlates of long‐distance economic interaction such as artifacts fashioned from exotic materials (Campbell, Morehead, et al. 2009; Thomas and Campbell 1991). Much of the recent discussion and refinement of the Elliott’s Point Complex is due to the work of Thomas, Campbell, and colleagues based on CRM‐related projects on Eglin AFB. 51
Bayou Park, 8OK898
Figure 2.9. Location of the Bayou Park site in relation to Poverty Point and other related sites.
Chronology Elliott’s Point has been subdivided into a three‐sequence chronology based on temporal changes in artifact assemblages (Campbell, Morehead, et al. 2009:17; Thomas and Campbell 1993:527‐528). The radiocarbon chronology and diagnostic artifacts or traits are listed in Table 2.2 and a list of relevant calibrated radiocarbon dates is presented in Table 2.3. A formative or “nascent” expression that began around 2500 BC is represented by sites such as 8OK898 and Meig’s Pasture (8OK102), located on the east side of Rocky Bayou (Campbell, Morehead, et al. 2009:17). This early phase of Elliott’s Point is characterized by smaller shell midden and accretional mound sites, a predominance of amorphous baked‐ clay objects, and minor evidence of trade represented by artifacts such as steatite vessels (Thomas and Campbell 1991). Radiocarbon dates from 8OK898 suggest the site was occupied from 2535 to 1860 BC, though the majority of ranges fall within 2500 to 2200 BC (Beta Analytic, Inc. 2002; Meyer et al. 2002:64). Dates from Meig’s Pasture place the site’s Elliott’s Point component at around 2425 to 1955 BC (Curren 1987:71‐73). 52
Environmental and Archaeological Contexts
Table 2.2. Elliott’s Point chronology and diagnostic traits. Age Diagnostic Traits 2500 ‐ ~2000 BC Small shell middens and accretional mound sites; amorphous baked‐clay objects; evidence of extra‐local trade (steatite). Classic ~2000 – 1000 BC Small campsites and larger “aggregation and redistribution” sites; “well‐ formed” baked‐clay objects, microliths, stemmed Archaic projectile points; exotic trade items; first appearance of fiber‐tempered pottery (late) Terminal 1000 – 600 BC Fiber‐tempered pottery; shift in shellfish consumption from quahog clam and scallop to Rangia SOURCE: Thomas and Campbell 1993; Campbell et al. 2004. Phase Nascent
A “classic” or fluorescent stage of the Elliott’s Point Complex is said to have followed sometime between 2000 BC and 1100 BC, typified by a greater number of exotic items, an increase in the number of “well‐formed” baked‐clay objects (Thomas and Campbell 1993:528), and evidence of long‐distance trade as well as trade redistribution through periodic population aggregation (Thomas and Campbell 1993:527‐528). A component of the Alligator Lake site (8WL29) on Fourmile Peninsula is considered a fluorescent‐age Elliott’s Point component, dating from 1675 to 1025 BC. Campbell et al. (2004:137‐139) also include a second occupation at Meig’s Pasture and the Graveyard Point site (8SR44) as fluorescent‐age sites. The former is said to date to 1700‐1455 cal BC while the latter, located at East Bay, is reported as dating to 1585‐1245 cal BC. However, the samples on which both of these dates are based are marine shell and it does not appear that any 13 12 C/ C adjustment was made to the measured radiocarbon ages of 3630 ± 50 BP and 3490 ± BP, respectively. When the corrected conventional radiocarbon ages are used, the calibrated dates are 2270‐1956 BC for Meig’s Pasture and 2124‐1729 BC for Graveyard Point. Several sites located on Fourmile Peninsula, including the Buck Bayou Mound (8WL90), which is believed to be a population aggregation site, are thought to have been most heavily utilized during Elliott’s Point fluorescence as well (Thomas, P. M. 1989; Thomas and Campbell 1991:104‐108, 1993:530‐532). The full range of the Elliott’s Point artifact assemblage developed during this period, including an increase in exotic materials presumably obtained through trade (Campbell et al. 2004:137‐140; Thomas and Campbell 1991:108‐113). The appearance of fiber‐tempered ceramics also occurred during the latter part of this phase, although they do not seem to have been widely accepted (Campbell et al. 2004). The final phase of the Elliott’s Point Complex is characterized by a series of events that occurred sometime after 1000 BC, including an increased acceptance of fiber‐tempered ceramics, a decrease in exotic trade items, and a shift in settlement patterns marked by greater use of interior, non‐coastal occupations (Campbell et al. 2004:139‐149; Thomas and Campbell 1993:528, 537‐540). It appears that this aspect of Elliott’s Point lasted until around 600 BC, after which the definitive attributes of the complex are replaced by those of the Deptford tradition (Campbell et al. 2004:140; Thomas and Campbell 1993:527‐528). 53
Table 2.3. Calibrated radiocarbon dates for Elliott’s Point sites. Site #
Site Name
Sample #
Material
8WL01278 8WL01278 8WL01278 8WL01278 8OK00898
Mitchell River 1 Mitchell River 1 Mitchell River 1 Mitchell River 1 Bayou Park
WK‐9644 WK‐9647 WK‐9651 WK‐9648 Beta‐171507
shell shell shell charcoal shell
8OK00898
Bayou Park
Beta‐129298
shell
8SR00017 8OK00898
Aden Bayou Bayou Park
Beta‐191432 Beta‐171508
8WL01005 8OK00102 8OK00102 8OK00102 8OK00102 8OK00102 8SR00017
NN Meig’s Pasture Meig’s Pasture Meig’s Pasture Meig’s Pasture Meig’s Pasture Aden Bayou
Beta‐81709 Beta‐21253 DIC‐3294 Beta‐21254 Beta‐21255 DIC‐b Beta‐203566
charcoal shell soot from steatite sherd shell shell shell shell shell charcoal
Mikell and Saunders 2007 Mikell and Saunders 2007 Mikell and Saunders 2007 Mikell and Saunders 2007 Meyer et al. 2002 Morehead, Campbell, and 3790±70 Aubuchon 2008 Thomas, Campbell, and b Mathews1994 3770±50 Meyer et al. 2002
‐1.4 ‐1.8 ‐0.9 ‐25.3 0.0
4280±50 4190±55 4145±50 3880±50 4210±60
Calibrated Years BCa (Confidence Level) 2581‐2285 (1.0) 2470‐2156 (1.0) 2432‐2121 (1.0) 2473‐2204 (1.0) 2529‐2178 (1.0)
0.0
4200±70
2532‐2135 (1.0)
b 0.0
b 4180±60
2470‐2210b 2468‐2132 (1.0)
3740±50 3700±80 3690±50 3670±80 3630±90 3630±50 b
‐25.3 ‐1.0 0.0d ‐1.0 ‐1.0 0.0d b
3735±50 4100±80 4097±50 4070±80 4030±90 4037±50 b
2291‐2014 (.98) 2437‐1976 (1.0) 2361‐2026 (1.0) 2402‐1933 (1.0) 2383‐1867 (1.0) 2270‐1956 (1.0) 2230‐1870b
0.0 0.0d ‐25.9 ‐25.0 b ‐25.0c ‐25.0c ‐25.0c ‐25.0c
3970±70 3897±70 3520±50 3390±80 3440±40 3135±125 3085±130 2670±150 2575±80
2228‐1817 (1.0) 2124‐1729 (1.0) 1976‐1737 (.98) 1888‐1502 (1.0) 1881‐1663 (.98) 1688‐1048 (1.0) 1631‐976 (.99) 1134‐406 (.97) 849‐481 (.93)
Measured 14C Years BP
13
Reference
b b b b 3800±50
Yates 2000 Curren 1987 Curren 1987 Curren 1987 Curren 1987 Curren 1987 Campbell, Morehead, et al. 2009 Morehead, Campbell, and 8OK00898 Bayou Park Beta‐129297 shell 3560±70 Aubuchon 2008 8SR00044 Graveyard Point Beta‐39718 shell 3490±70 Thomas & Campbell 1993 8WL01278 Mitchell River 1 WK‐9689 charcoal b Mikell and Saunders 2007 8WL01278 Mitchell River 1 Beta‐139437 charcoal b Mikell and Saunders 2007 8WL02233 NN not identified charcoal b Campbell, Bourgeois, et al. 2009 8WL00029 Alligator Lake FSU‐32 charcoal 3135±125 Lazarus 1965 8WL00029 Alligator Lake FSU‐64 charcoal 3085±130 Lazarus 1965 8WL00035 Four Mile Village b charcoal 2670±150 Lazarus 1965 8WL00029 Alligator Lake GX‐0155 charcoal 2575±80 Lazarus 1965 a Calibrations performed using CALIB 6.01 and INTCAL09, MARINE09 calibration datasets (Stuiver and Reimer 1993). b Data not reported. c 13 12 C/ C ratio not reported; used standard estimate of ‐25.0‰ for charcoal (Stuiver and Polach 1977). d 13 12 C/ C ratio not reported; used standard estimate of 0.0‰ for marine shell (Stuiver and Polach 1977).
C/12C o/oo
Conventional 14 d C Years BP
Environmental and Archaeological Contexts
Sites occupied during the Deptford period include Alligator Lake (8WL29), with an occupation radiocarbon dated to between 840 and 415 BC (Campbell et al. 2004:137‐139). The factors contributing to the decline of the Elliott’s Point Complex, and the emergence of the subsequent Deptford culture, are not well understood. Morehead et al. (2011:24‐25) suggest that it may have been tied to the roughly contemporaneous demise of Poverty Point, which has itself been hypothesized as resulting from massive flooding of the Lower Mississippi Valley as a result of global cooling and increased precipitation (Kidder 2006; but see Kidder 2010 for a more nuanced interpretation of Poverty Point’s demise). An alternative explanation, offered by Ken Sassaman (1993:225‐227), is that trade in raw materials, so important to the rise of Poverty Point, was disrupted by the widespread adoption of ceramic pottery from the east, while Gibson (2010) suggests that the people of Poverty Point may have returned to a mobile lifestyle in order to facilitate communication and social networking between related groups. At present, the question of how and why Elliott’s Point (and Poverty Point) disappeared from the archaeological record approximately 2500 years ago remains unresolved.
Material Culture According to Thomas and Campbell (1991:108‐113; 1993:532‐537), the artifacts that identify Elliott’s Point archaeologically include baked‐clay objects, microlithic tools (often referred to as Jaketown Perforators), exotic trade items (primarily steatite), and Florida Archaic Stemmed projectile points, particularly a type known as Destin (Thomas and Campbell 1991:110‐111; Campbell, Morehead, et al. 2009:17). Baked‐clay objects around Choctawhatchee Bay are found in a variety of sizes and may be “well‐formed” or amorphous in shape and can be plain or exhibit surface treatments and decorations. Amorphous objects are the most numerous (McGee 1995:104; Thomas and Campbell 1993:533). Steatite, in the form of stone vessels, pipes, boatstones, and ornaments, is also common in both early and late Elliott’s Point assemblages (Campbell, Morehead, et al. 2009:17‐18; Thomas and Campbell 1993:534). Thomas and Campbell (1991:108) note an absence of the lapidary artifacts so common to Poverty Point‐related assemblages elsewhere (e.g., hematite plummets); they surmise that the few specimens that have been recovered, such as a jasper gorget from 8WL87 and a jasper figurine from the shore of Choctawhatchee Bay, were obtained through trade. A “moderately active” bone and shell industry is also considered a characteristic of Elliott’s Point material culture, including the manufacture and use of bone tools as well as bone and shell beads (Thomas and Campbell 1991:108). Although microliths are common, not all resemble Jaketown Perforators, as Fairbanks (1959:97) pointed out. Fiber‐tempered pottery of the Norwood series is considered a late addition to an already well‐established Elliott’s Point cultural tradition (Campbell et al. 2004; Thomas and Campbell 1991, 1993), and its arrival had no observable impact on the economic and technological characteristics that defined the Elliott’s Point Complex (Campbell et al. 2004:149). Fiber‐tempered ceramics are estimated to have been incorporated sometime 55
Bayou Park, 8OK898
between 1675 and 1025 BC based on data from Alligator Lake, though the possibility of earlier incorporation is acknowledged (Bourgeois et al. 2009:19; Campbell et al. 2004:139). Campbell, Bourgeois, et al. (2009:96) report a much earlier date of 2500 to 2300 BC from a feature at 8WL2233 where two sherds of fiber‐tempered pottery also was found; however, the sherds were not recovered from the feature but from approximately 40‐50 cm above the top of the feature (Campbell, Bourgeois, et al. 2009:81, 87). Thus the association of fiber‐tempered ceramics with the radiocarbon date is debatable. There also appears to be an issue with the reported calibration (cf. Table 2.3). The earliest dates for fiber‐tempered pottery then are from Alligator Lake and these are at least 1,500 years later than the acceptance of pottery elsewhere in Florida and the rest of the region (White 2003a, 2004). Assuming that fiber‐tempered pottery did not appear in the Choctawhatchee Bay region until relatively late, the logical question is why ceramic technology was resisted here long after it was accepted elsewhere? The answer may be related to the socioeconomic forces influencing Elliott’s Point society. Sassaman (1993) suggests that ceramic vessel technology diffused slowly and erratically westward after its introduction on the east coast. He suggests that part of the reason that ceramic technology was so slowly accepted, was that it undermined the value of steatite vessels. The groups and individuals who controlled the Late Archaic trade networks likely enjoyed prestige and power, as well as the ability to influence the pace and direction of technological innovation. A technology that allows for the creation of vessels from local clay sources would have rendered the steatite vessels obsolete, and as steatite was likely a valuable market within the trade network, the magnates who benefited most from the movement of steatite may have actively campaigned against the inception of ceramic vessel technology. Similar examinations of the socioeconomic relationships between steatite and ceramic vessels in Georgia have also been undertaken by Wagoneer (2009). Campbell et al. (2004:148) suggest that trade in steatite was a significant part of the local Elliott’s Point economy, resulting in an economic aversion to ceramic vessels and its rather late and limited acceptance. Considering the commonality of steatite in Elliott’s Point assemblages (Thomas and Campbell 1993:534), this interpretation seems valid if untested.
Settlement Patterns Most Elliott’s Point sites are located in coastal settings, particularly around bays, with a minor occurrence of sites beyond 1.5 km from the bay (Thomas and Campbell 1991:104, 1993:528). Calvin Jones (1993) documented clusters of Elliott’s Point sites around Choctawhatchee, Apalachicola, St. Andrews, Apalachee, and Pensacola bays, with Choctawhatchee Bay having the greatest number of sites. Sites there tend to cluster around virtually every bayou on all sides of the bay and around specific locations such as East Bay, Cinco Bayou and the Narrows, Boggy and Rocky Bayous, Fourmile Peninsula, and Basin Bayou (Figure 2.10). Of the interior sites that have been recorded, favored locations are on the terraces and ridges overlooking rivers or at the heads of tributary seeps (Thomas and Campbell 1993:256). These distributions suggest a heavy reliance on ecotonal coastal locations with potable water that primarily occur along the shores of the bay proper and along the bayous and creeks that drain into the bay. 56
Figure 2.10. Distribution of Late Archaic sites in the Choctawhatchee Bay region. Data from Florida Master Site file (2012) and Campbell et al. (2004). Base Map: SAIC (2010)
Environmental and Archaeological Contexts
57
Bayou Park, 8OK898
The types of Elliott’s Point sites that are present around Choctawhatchee Bay are not well understood. Among those that have been identified are smaller sites referred to as stations (e.g., 8OK2694, Mathews et al. 2009), exploitation sites (e.g., 8WL173, Campbell et al. 2007), or resource extraction camps (e.g., 8WL2270, Mallory and Campbell 2008). These smaller sites are typically located near interior drainages and have no associated middens or features. Lithic workshops and specialty craft sites, defined by a concentration of lithic tools, debitage, and microdrills, also have been identified (Thomas and Campbell 1993:531‐ 532). Other sites include occupational loci referred to as encampments (e.g., 8WL36, Thomas et al. 2001), seasonal camps (e.g., 8OK427, Campbell et al. 2008), or short‐term camps (e.g., 8WL2126, Mallory and Campbell 2006). These sites typically have middens and features and are often located in clusters, perhaps to take advantage of nearby aggregation or “redistribution” centers (Thomas and Campbell 1993:528). Shell middens/mounds that represent more extensive occupation sites, regional centers, or potential population aggregation sites are also present. These sites are usually characterized by circular or horseshoe‐shaped (arcuate) shell deposits. Alligator Lake (8WL29) is thought to be one of these prominent habitation sites, while Buck Bayou Mound (8WL90) may potentially represent an arcuate feasting monument associated with trade redistribution (Thomas 1989; Thomas and Campbell 1993:529‐532). The Buck Bayou Mound is unique among Elliott’s Point sites in part because of its size, nearly two meters high with a basal dimension of 125 by 110 m (Thomas 1989:4). While there is no evidence that the mound was used for human interments, it contains exotic materials such as steatite and quartz tools, as well as bone tools, cut shell beads, and fragments of baked‐clay objects. Meig’s Pasture (8OK102), an arcuate shell midden, is interpreted as having been a seasonal camp rather than a redistribution center due to the lack of “exotic” artifacts and its small size (Thomas and Campbell 1993:528‐529). However, Campbell et al. (2009a:19) acknowledge the possibility that Meig’s Pasture may be a predecessor of Buck Bayou Mound for early Elliott’s Point population aggregation. Thomas and Campbell (1991:104‐105, 1993:528) assert that Elliott’s Point settlement patterns are very similar to those of Poverty Point‐related sites in the Mississippi Valley, where smaller occupation and specialty sites tend to cluster around larger sites thought to have been regional centers of distribution and influence. This pattern is represented most directly at Fourmile Peninsula, where occupation sites (e.g., 8WL36) and specialty lithic sites (e.g., 8WL192) are located in close proximity to the Buck Bayou Mound. However, the majority of Elliott’s Point sites identified around Choctawhatchee Bay have been identified as camps rather than “extensive” habitations (Thomas and Campbell 1993:528). It appears that a shift in settlement systems may have occurred toward the end of the Elliott’s Point Complex that coincided with both a greater utilization of fiber‐tempered ceramics and a drop in the salinity levels of Choctawhatchee Bay (Campbell et al. 2004:140; Thomas and Campbell 1991:114‐115, 1993:539‐549). Fiber‐tempered‐using groups, thought to have become prominent by at least 1100 BC (Thomas and Campbell 1993:527‐ 528), exhibited new settlement preferences for the shores of brackish water bayous and for 58
Environmental and Archaeological Contexts
interior occupations along transportation routes following major and minor streams (Campbell et al. 2004:140). In addition, sometime around 1000 BC, the slow accretion of the barrier island forming Moreno Point likely restricted water exchange between the Gulf and the bay. As discussed above, this event may have resulted in lower salinity levels and a reduction in suitable habitat for quahog and bay scallop, which formed a significant part of the Elliott’s Point diet. This desalination of the bay may have contributed to the settlement shifts noted for this time and perhaps contributed to the onset of the succeeding Deptford period (Campbell et al. 2004).
Subsistence and Seasonality Although the triad of oyster, quahog, and bay scallop are sometimes referred to as diagnostic of Elliott’s Point subsistence (e.g., Thomas and Campbell 1993:541), there are few detailed data on subsistence and seasonality practices. Elliott’s Point subsistence discussions are typically limited to gross shellfish proportions and descriptions of vertebrate fauna recovered from large‐mesh (1/4‐inch) screens (e.g., Campbell et al. 1998; Meyer et al. 2002; Thomas 1989; Thomas, Campbell, Mathews, and Morehead 1994). An exception is Meig’s Pasture where a large array of invertebrate and vertebrate taxa was recovered using small‐mesh screens. The analysis of these samples indicates extensive use of aquatic faunal resources (Russo 1987). In addition to bay scallop, quahog clam, and oyster the remains of Rangia, marsh periwinkle, ray, shad, sea catfish, jack, white‐tailed deer, turtle, rabbit, and river otter were also recovered. Another well‐documented Late Archaic site is Mitchell River 1 (8WL1278) on the east side of Choctawhatchee Bay (Mikell and Saunders 2007). This site contains an Elliott’s Point component; however, the faunal data for all of the site’s Late Archaic components were collapsed into a single table so it is difficult to characterize Elliott’s Point subsistence on its own. It is worth noting, however, that a variety of mammals, reptiles, and a few freshwater fish were recovered from excavations at the site, in addition to marine fish and shellfish remains (Mikell and Saunders 2007:Table 6). Bayou Park has been interpreted as having been an early Elliott’s Point seasonal camp with evidence of varied activities (Campbell, Morehead, et al. 2009:60). Campbell et al. (1998:70) conclude that, typical of early Elliott’s Point sites, the dominant shellfish species are oyster, quahog clam, and bay scallop. Feature 3 is recorded as having been dominated by oyster and quahog with “appreciable” amounts of conch and only a minor incidence of bay scallop (Campbell et al. 1998:70). The midden at 8OK898 consisted of primarily oyster, bay scallop, and quahog with some Rangia, Strombus, and moon snail. Limited vertebrate remains indicate that shellfish was augmented with opossum and deer, fish (all were unidentified), turtle, and bird; however, the vertebrate fauna is biased towards larger elements since the sample was derived from large‐mesh recovery screens. Consequently, fish remains are underrepresented. Seasonality studies at Elliott’s Point sites are limited. The most detailed studies have been conducted for the Meig’s Pasture site. Claasen (1987) analyzed shellfish size and internal shellfish growth ring data at Meig’s Pasture, but the results of the analyses were 59
Bayou Park, 8OK898
contradictory (Claassen 1987). The size of analyzed specimens suggested collection during warm months, while internal growth ring data implied collection during periods of cooler water temperatures. Claassen (1987:43) suggested that, given differences in the sample sizes, more validity should be given to the internal growth ring data, which suggest cool weather occupation. Hickory nut shells were recovered in abundance, and a single grape seed and a cabbage palm seed were also recovered (Newsom 1987:45‐49). It is likely that the hickory nuts were utilized for subsistence, but the grape and cabbage palm occurred in such low numbers that it is difficult to arrive at any conclusion regarding their use as food. The presence of hickory mast further suggests a fall or winter occupation. Russo (1987:56‐57) observed that most of the fish species in the sample are found in shallow‐water estuarine environments and, in general, are most common during warm months. Based on these data (of which the sample size was much larger than the plant remains), he concludes that Meig’s Pasture was occupied during a warm‐weather season (Russo 1987:57). In retrospect, however, the shellfish, botanical, and faunal data seem to contradict one another in terms of a single season of occupation. The most parsimonious explanation is that Meig’s Pasture was occupied during several seasons of the year, although a more thorough seasonality study is probably warranted. Hickory nut shells were recovered from Bayou Park during PTA’s Phase III investigation in 1998, indicating that the inhabitants were engaged in gathering activities at least during portions of the fall and perhaps winter (Campbell et al. 1998:75). This may suggest that the site was occupied either during fall or winter months, or that hickory mast was stored and processed at 8OK898 out of the season of harvest. Additional information on Elliott’s Point seasonality can be inferred from the Buck Bayou Mound (Thomas 1989; Thomas and Campbell 1993). The depositional structure of the mound strata suggest that the site was not occupied year‐round, but rather represents occasional, high intensity visits punctuated by periods of abandonment (Thomas and Campbell 1993:531). In addition, the presence of bay scallops is interpreted as representing a summer occupation, as bay scallops enter the shallow waters of Choctawhatchee Bay primarily during the warm months, though Thomas and Campbell (1993:531) admit that this evidence is not conclusive. However, if Buck Bayou Mound represents a periodic aggregation site, as they claim (Thomas and Campbell 1993:529‐531), rather than a seasonal camp or village, then these data may reflect seasonal patterns of population aggregation rather than the seasonal settlement movements of one or a few groups. Indirect evidence of seasonal movements by Elliott’s Point people has also been recorded at 8WL1005 on Alaqua Creek, a watercourse that feeds into Choctawhatchee Bay, where a cache containing two steatite bowls was found (Hemphill et al. 1995). The caching of heavier site furniture may denote the desire to travel unencumbered, as well as the intention of returning to the cache area to retrieve the items later. In sum, very little of a detailed nature is known about Elliott’s Point subsistence around Choctawhatchee Bay, and by extension, about the scheduling of Late Archaic seasonal 60
Environmental and Archaeological Contexts
movements. The majority of sites around the bay have been identified as seasonal “camps,” and subsistence appears to have been centered on marine invertebrate species. However, this latter interpretation is based on limited samples collected with large‐mesh screens and do not adequately reflect the full range of maritime subsistence, specifically the importance of fish, whose small bones escape detection without the use of small‐mesh screens. Where detailed analyses have been undertaken it is apparent that shellfish were augmented with a variety of other food sources, including fish and several terrestrial species.
Trade in the Choctawhatchee Bay Region From Choctawhatchee Bay, Elliott’s Point populations would have had access to both bountiful resources and avenues of transportation to distant groups and their resources. The evidence usually cited for participation in long‐distance exchange includes the presence of items fashioned from foreign materials such as exotic cherts, steatite, copper, jasper, galena, hematite, and even pumice (Campbell, Morehead, et al. 2009:18‐20; Thomas and Campbell 1991:112‐113, 1993:531‐533). Fragments of steatite vessels have been found at the Buck Bayou Mound, Horseshoe Bayou (8WL36), Pirate’s Bay (8OK183) and caches of steatite vessels have been recovered from 8WL87 and 8WL1005 (Gagliano and Webb 1970:71; Thomas and Campbell 1993:534). Diagnostic artifacts common in the assemblages of contemporary groups throughout the region, such as baked‐clay objects, are also abundant in many Elliott’s Point sites, presumably connecting local populations to groups much further away. Other lines of evidence include similarities in projectile point styles. Seven projectile points with straight‐sided stems were recovered from the Alligator Lake site and are described as Delhi points, a style that was popular at Poverty Point (Lazarus 1965:95, 98). Delhi points are corner‐notched with drooping barbs (Ford and Webb 1956:59‐60). Only five of the Alligator Lake specimens are pictured but only two resemble the type description for Delhi points. The other three points display straight basal blade corners. Lazarus also describes two basally notched points as Marshall points, another projectile point type that was found at Poverty Point. One of the Alligator Lake specimens reportedly is made of clear quartz crystal (Lazarus 1965:98). Gagliano and Webb (1970) state that aspects of the Buck Bayou Mound assemblage compare closely with items found at the adjacent Cedarland and Claiborne sites, located at the mouth of the Pearl River on the Mississippi Gulf Coast (see Figure 2.9). Both of these sites are believed to have been directly associated with the Poverty Point Exchange Network (Gagliano and Webb 1970; Russo 1996). A cache of 10 steatite vessels was recovered from the Claiborne site (Gagliano and Webb (1970:58‐59) and projectile points similar to the Florida Archaic Stemmed Levy subtype were found at both sites (Gagliano and Webb 1970:59). The construction of arcuate shell middens at Elliott’s Point sites is similar to the arcuate middens at Cedarland and Claiborne. 61
Bayou Park, 8OK898
Connections with Poverty Point‐related sites presumably occurred primarily along the coast. In this interpretation, Alligator Lake was the initial contact point for long‐distance Poverty Point traders because of its proximity to the Gulf of Mexico and inland‐reaching watercourses. The Buck Bayou Mound on Fourmile Peninsula served as an aggregation site for Elliott’s Point villages in the surrounding area (Campbell et al. 2009a:19; Thomas and Campbell 1993:532), facilitating large feasting events and inter‐group trading that ultimately resulted in the redistribution of goods as well as the accretion of a mound that was likely recognized as a monument to fellowship and the prestige of the hosting parties (Russo 2004). This interpretation is based partly on the stratigraphic interpretation of the Buck Bayou Mound itself, which is interpreted as representing large, episodic feasting events punctuated by periods of abandonment (Thomas and Campbell 1993:532). With the “decline” of the Elliott’s Point Complex, involvement in long‐distance exchange appears to have been reduced significantly. The late acceptance of fiber‐tempered ceramics coupled with a decrease in exotic items suggests both a reduction in exchange activities and a waning of the influences of the trading elite that favored steatite. Indeed, the Deptford culture that followed Elliott’s Point exhibits few signs of long‐distance interaction and has been described as “conservative” in comparison to the cultures of the Late Archaic period (Campbell et al. 2004:149). Further, it may be no coincidence that a reduction in Elliott’s Point trade and the influx of ceramic vessels roughly coincide with the drop in bay salinity that resulted in changes in shellfish species composition and possible shifts in settlement patterns (Campbell et al. 2004:148‐149). Although there are several scenarios that could explain the correlations between the acceptance of ceramics, changes in subsistence availability, shifts in settlement patterns, and shifts in the local power structure, the manner in which these events are connected remains uncertain.
Mounds, Ceremonialism, and Social Interaction Ceremonialism, mortuary practices, and mound function are poorly understood in the Choctawhatchee Bay area as they are across the Late Archaic Southeast (Gibson 2000; Thomas and Campbell 1993). Elliott’s Point mounds are nearly exclusively shell and are only found in coastal settings. These include Buck Bayou Mound, Graveyard Point, and Meig’s Pasture. Russo (2004:40) suggests that shell found in piles, lenses, and other features associated with shell mounds represent the epiphenomena of interactions, particularly of feasting events. He argues that with its open plaza and raised circular boundary, the arcuate shell structure provided a public forum for all aspects of food preparation, distribution, and disposal (Russo 2004:41). These public rituals facilitated formal social interactions and affirmed social connections. The structured, conspicuous disposal of the shell produced by these events served as the ceremonial seat of rituals associated with the feast activities and memorialized the host’s generosity and prestige. Thus, it is possible to interpret Elliott’s Point shell mounds as representing feasting monuments resulting from population aggregation whereby ritual gathering was intended to reaffirm economic and social ties while simultaneously redistributing wealth obtained through exchange interactions within the Poverty Point Exchange Network. 62
Environmental and Archaeological Contexts
The identification of Meig’s Pasture as an intentionally constructed or accretional mound has been questioned, however, by Caleb Curren who conducted excavations at this site in 1987 (Curren 1987:11‐15). On the basis of several trenches excavated through the “mound” Curren concluded that the site consists of numerous shell deposits and pit features that were situated on a low sandy rise surrounding a shallow depression that may have held water at one time (Curren 1987:78). His trench profiles clearly show the discontinuous and overlapping nature of the deposits and features, and support his interpretation that is not an accretional midden or intentionally constructed tumulus, but is the result of periodic refuse disposal events. This does not invalidate the claim that the site may have served as a locus for feasting, however, nor does it contradict the suggestion of Thomas and Campbell (1993:528‐530) that Meig’s Pasture may have been a precursor of the accretional mound building and population aggregation events that culminated in the Buck Bayou Mound. Unfortunately, there are no radiocarbon dates from Buck Bayou Mound or the surrounding midden with which to establish temporal relationships between the two sites. The only sand mound that is of potentially associated with Elliott’s Point is 8WL10 (known as Mound near Mack Bayou), located on the south shore of Choctawhatchee Bay just east of Fourmile Peninsula. Moore (1918:541) reported a stone pendant in association with a soil discoloration that may have been a badly decomposed burial. An “arrowhead or knife, of flint,” a sandstone hone, and a “bar amulet” were also recovered from the mound. Willey (1949:221) suggested the possibility that 8WL10 may have been a preceramic burial mound but the evidence was inconclusive. Thomas and Campbell (1993:529) conclude that, as mound burial is not a definitive attribute of Poverty Point‐related cultures, the mound’s relationship to the Elliott’s Point Complex is questionable. Recent research, however, has revealed that there is a great deal of variability in the burial practices of contemporary Late Archaic populations, including instances of mound burial (Russo 1994). Additionally, sand mounds have been associated with Late Archaic populations across the region (Peacock et al. 2010) and 8WL10’s association with the Elliott’s Point Complex may need to be reconsidered. At Mitchell River 1 (8WL1278) near the mouth of the Choctawhatchee River, 10 burials were discovered in two locations on the western end of a low (ca. 40 cm) rise that may be the remains of a small mound (Mikell and Saunders 2007:178‐179). The burials were interred in shell‐filled pits or shell midden deposits. The pits intruded into the underlying sand with the tops of the burials varying from 62 to 92 cmbs. Two radiocarbon dates were obtained for the burial area: 4840‐4570 cal BP from Level 8 (70‐80 cmbs) and 5900‐5660 cal BP from Level 9 (80‐90 cmbs). Since the burials are intrusive into earlier deposits, at least some of the shallower burials post‐date this time frame. Other portions of the site have been dated to the Elliott’s Point period (see Table 2.3) and amorphous baked‐clay objects and a steatite vessel rim sherd also were recovered.
63
Bayou Park, 8OK898
With the possible exceptions of 8WL10 and 8WL1278, no definitive Elliott’s Point burials have been discovered to date. Considering the level of work that has been conducted at Elliott’s Point sites, the absence of burial contexts is potentially significant in terms of associations with contemporaneous groups. The absence of a shared mortuary complex may denote a lack of ideological associations between Elliott’s Point populations and contemporaneous Late Archaic groups, or at least the lack of evidence for such a connection.
Post‐Archaic in the Panhandle
Woodland Period The widespread use of ceramics, the interment of the dead in mounds, and increased regionalization characterize the Woodland period, which began at around 2600 BP (600 BC) in the Choctawhatchee Bay region and continued until about 1000 BP (AD 1000) (Campbell et al. 2004:140; Thomas et al. 1996). The first archaeological manifestation that is recognized as a Woodland culture is Deptford. Stephenson et al. (2002:325‐333) define three regional variations in the Southeast: Gulf, Atlantic, and Coastal Plain‐Interior Riverine; Florida’s panhandle falls within the Gulf region. Florida Deptford is described as primarily a “coastal dwelling culture” which relied heavily on maritime subsistence strategies (Milanich and Fairbanks 1980:66). Deptford villages located on the coast are usually found in conjunction with live oak, magnolia, and palm hammocks located near salt marshes. Interior Deptford is found along lakes and streams where hickory and oak are present. These are believed to represent seasonal camps, presumably occupied during the late fall and winter. Deptford sites contain plain pottery or ceramics with checked patterns stamped on the exterior of the pot, and many have distinctive podal supports (Milanich 1994). Temporally, two phases have been identified in the Choctawhatchee Bay region: Alligator Lake (ca. 600‐50 BC) and Okaloosa (50 BC‐AD 150) (Thomas and Campbell (1993:547‐554; Thomas et al. 1996). A possible third phase was identified at the Early Fish Fry site (8OK126) overlying Alligator Lake phase deposits and producing radiocarbon dates of 330 and 320 BC (Thomas and Campbell 1993:257). Excavations at Pirate’s Bay (8OK183) have produced important information about Deptford subsistence and settlement (Thomas et al. 1991). This coastal site has produced evidence that Deptford people exploited a wide range of local marine and terrestrial food resources. In addition, trade items from the Lower Mississippi Valley and southern Georgia were recovered, providing evidence of participation in a far‐flung exchange network with neighboring cultures. The lithic assemblage is distinguished by the presence of small, backed white quartz pebbles that appear to have been specialized tools.
64
Environmental and Archaeological Contexts
An important component of Deptford culture in northwest Florida is a mortuary‐ceremonial complex referred to as Yent and originally defined by Sears (1962). The Yent Complex included the inclusion of exotic goods obtained through exchange in mortuary contexts. These exotic items, which included galena, mica, artifacts made of metamorphic rock, and similarities in ceramic vessel design, may have come to Florida via exchange with Hopewell cultures farther north (Milanich 1994). The Yent Complex appears strongest in the Big Bend region of northwest Florida. Farther west, burial mounds and characteristic Yent Complex artifacts are rare (Milanich 1994). Instead, Deptford people disposed of their dead in graves within or adjacent to their villages (Thomas and Campbell 1993). The Santa Rosa‐Swift Creek culture replaced Deptford throughout northwest Florida, beginning about 1850 BP (AD 150). It is marked by two ceramic series: Swift Creek and Santa Rosa. It seems that Swift Creek ceramic designs originated in southern Georgia and were subsequently adopted by Deptford people in Florida. Swift Creek pottery exhibits complicated stamped designs consisting of scrolls, concentric circles, teardrops, and spirals. Check stamping also was used by Swift Creek potters. Swift Creek vessel forms include squat bowls and deep cylindrical pots. The replacement of Deptford ceramics by Swift Creek in the Florida panhandle took place over several centuries. Santa Rosa ceramics contain incised, punctated, and rocker‐stamped designs and appear to be a continuation of ceramic traditions that originated in the Lower Mississippi Valley. Santa Rosa‐Swift Creek villages were located on the coast and in the interior forests and river valleys throughout the panhandle. Excavations at Horseshoe Bayou (8WL36) and the Old Homestead site (8WL58) have provided information about Santa Rosa‐Swift Creek subsistence, settlement, and socio‐political and religious organization (Thomas et al. 1996, 1998, 2001). 8WL58 consists of a circular shell midden surrounding a plaza. Rangia accounts for 99% of the shell from the site. The plaza contains no shell midden, but a dark earth midden was identified and excavation revealed numerous postmolds. At 8WL36 a horseshoe‐shaped shell midden composed of Rangia is present. Analysis of faunal remains indicated that a wide range of local marine and terrestrial food resources were exploited (DeFrance 2001) Based on their work at these sites, Thomas et al. (1996) defined two phases for Santa Rosa‐ Swift Creek in the Choctawhatchee Bay region: Lassiter (ca. AD 150‐450) and Horseshoe Bayou (ca. AD 450‐650). The former is characterized by Swift Creek Complicated Stamped pottery, St. Andrews Complicated Stamped, West Florida Cord Marked, Crooked River Complicated Stamped, Alligator Bayou Stamped, Santa Rosa Stamped, Basin Bayou Incised, Gulf Check stamped, and Franklin Plain. The Horseshoe Bayou phase is distinguished by a distinctive ceramic type – Horseshoe Bayou Complicated Stamped – that exhibits a bold check stamp with a raised dot in the center of the check. The socio‐religious aspect of this culture has been defined as the Green Point complex (Sears 1962), which had associations with the Hopewell interaction sphere and may have 65
Bayou Park, 8OK898
developed out of the preceding Yent Complex. Through this exchange network, Santa Rosa‐ Swift Creek people gained access to exotic items, such as copper, mica, ear spools, and ceramics (Bense 1998). The emergence of Weeden Island cultural attributes in the Choctawhatchee Bay region began at about 1350 B.P. (AD 650). Weeden Island has been defined as a religious‐ ceremonial complex that was adopted by local regional cultures in southern Georgia and Alabama and along the west coast of Florida (Milanich 1994). Early Weeden Island is characterized by the appearance of complicated stamped pottery along with the characteristic Weeden Island pottery decorated with incised and punctated lines (e.g., Carrabelle Incised, Carrabelle Punctated, Keith Incised, and Weeden Island Incised). There appears to be some continuity between Santa Rosa‐Swift Creek and Weeden Island occupations. Not only are both cultural expressions found in the same coastal environmental settings, but these cultures exploited similar marine resources. Fish remains include herring, saltwater catfish, sea catfish, jack, porgies, sheepshead, mullet, flounder, bowfin, drum, and gar. Shell middens indicate a preference for oysters, although conch, Rangia and other species also are present. Vertebrate faunal remains include white tail deer, freshwater turtle, and birds. Acorns and hickory nuts were collected as were various plant species, such as yaupon, wild grape, palmetto shoots, and gallberry. Mortuary ceremonialism reached its peak during early Weeden Island times. Ornately decorated ceramics and those shaped as stylized designs or animal effigies were interred in burial mounds, often on the mounds’ east side (Milanich and Fairbanks 1980). Early Weeden Island villages also appear to have been arranged in circular patterns as evidence by several “ring” sites that were identified on the Tyndall Air Force Base (Russo et al. 2009). Late Weeden Island (AD 750‐1000) is identified by the presence of check‐stamped and cob‐ marked pottery, and is referred to as Wakulla Weeden Island by Milanich (1994:194‐204). Wakulla sites are located on the coast and in the interior of the panhandle, as well as in southwest Georgia and southeast Alabama. According to Milanich (1994:194), maize agriculture was adopted in the panhandle during the Wakulla period; however, no evidence of maize has been documented for the Choctawhatchee Bay region. Interior Weeden Island sites in this region tend to be smaller and distributed around springheads, a trend suggested by Milanich and Fairbanks (1980) as distinctive of the culture.
Mississippi Period For this report, the term “Mississippi” is used for the chronological period beginning about 1000 years ago (ca. AD 1000) and continuing until about 450 BP (AD 1550), while the term “Mississippian” is reserved for the broader late prehistoric/protohistoric cultural tradition that emanated from the Mississippi Bottoms and influenced regional cultures such as Pensacola and Ft. Walton. These influences included maize agriculture, shell‐tempered pottery, institutionalized social inequality, a chiefdom level of political organization, and participation in long‐distance exchange relations that involved the movement of exotic items and religious iconography throughout the Southeast (Ashley and White 2012). 66
Environmental and Archaeological Contexts
The Fort Walton culture was centered in the Red Hills surrounding modern Tallahassee and extended west to Choctawhatchee Bay. It was only in the Red Hills, however, that most of the trappings of the wider Mississippian phenomenon, including maize agriculture, were adopted. Farther west around Pensacola Bay, the Pensacola variant of the Fort Walton culture evolved in place out of the preceding Weeden Island culture and was influenced by cultural developments in Alabama (Harris 2012; Milanich 1994). The Choctawhatchee Bay region is interesting because ceramic series identified with both of these variants are present. The Fort Walton series is characterized by grit‐ and/or sand‐tempered pottery with distinctive incised and punctated surface designs as well as undecorated vessels (Willey 1949:452‐488). The Pensacola series is distinguished from the Fort Walton series by the presence of shell tempering associated with both decorated and plain vessels (Willey 1949). Two Mississippi‐period phases have been identified for the Choctawhatchee Bay region based on the increasing frequencies of Pensacola series pottery in Late Fort Walton sites (Mikell 1990, 1992). The Indian Bayou phase (ca. AD 1000‐1350) is dominated by Fort Walton series pottery with small frequencies of Pensacola series sherds. The Four Mile Point phase (ca. AD 1350‐1600) is characterized by relative frequencies of Pensacola pottery that range from between 30 and 40 percent to as much as 70 percent of the collections. Mississippi‐period villages are located almost exclusively in coastal locations. Few interior village sites are known and most of these are along the Yellow River, at the headwaters of south‐flowing tributaries, or on creeks at settings inland from Choctawhatchee Bay (Thomas et al. 1996). Major villages were likely occupied year‐round by at least limited populations, while the smaller hunting, gathering, and horticultural loci were occupied seasonally by small groups. The principal village pattern is represented by 8WL51, an off‐Eglin site on the west side of Hogtown Bayou, which Lazarus (1971:45) describes as “... six or seven small midden piles of shell ... arranged in a pattern.” Smaller Mississippian coastal sites were less intensively utilized and non‐nucleated, and may represent dispersed households and resource exploitation or special function campsites. Although there were fewer mounds than during the preceding Weeden Island period, there is clear evidence of ceremonialism in Fort Walton/Pensacola culture. Thomas et al. (1996) indicate that at least six Mississippian mounds exist in the Choctawhatchee Bay area and that these contain a variety of Fort Walton and Pensacola ceramics. The Fort Walton Temple Mound (8OK6) is the most impressive of these. This large platform mound measures 12 ft in height, 223 ft by 220 ft at its base, 90 ft by 150 ft at its summit. Over 80 burials are reported to have been interred in the mound and it may have served as a regional center of Fort Walton/Pensacola activity. The site has been the subject of several investigations (Fairbanks 1965; Lazarus and Fornaro 1975) that have produced evidence of multiple burials, shell and bone tools, shellfish, and vertebrate fauna, lithics, and mica. 67
Bayou Park, 8OK898
CONCLUSION In this chapter, we have attempted to summarize what is known about the environment and archaeology of the Choctawhatchee Bay region in order to provide a context for what follows. The focus has been on the Late Archaic period and the Elliott’s Point Complex because the principal occupation of the Bayou Park site was during this archaeological period. Special emphasis was placed on describing what is known (and what is not) about the people who inhabited this part of the Florida panhandle approximately 4500 to 2500 years ago. This information, and particularly the gaps in our knowledge about the Elliott’s Point complex, was used to develop a research design that is the subject of Chapter 3.
68
3
RESEARCH DESIGN AND METHODS
Archaeological work conducted over an 18‐year period at Bayou Park (8OK898) has demonstrated that the site’s primary occupation was during the Elliott’s Point phase of the Late Archaic period, with only minor evidence for later Weeden Island and Fort Walton components (see Chapter 1). A historic‐period (mid‐twentieth century) component also is present. Unlike other sites where overlying occupations are extensive and separation of the Elliott’s Point component is sometimes difficult, Bayou Park represents a nearly “pure” Elliott’s Point site with subsurface features, a midden containing preserved faunal and botanical remains, and lithic, ceramic, and bone artifacts. The site is eligible for the NRHP because these intact prehistoric deposits have potential to contribute to a better understanding of the Elliott’s Point Complex. Consequently, the primary focus of the data recovery effort was on gathering data related to the Late Archaic, Elliott’s Point component at 8OK898 in order to mitigate impact from future site development to this NRHP‐eligible component. The summary of previous work in Chapter 1 indicates that much valuable information has been obtained from the site, particularly in terms of general site layout, documentation of the stratigraphic and spatial distributions of major artifact classes, and basic chronology. The approach taken for this data recovery effort was to utilize as much existing information as possible to avoid duplication of effort and to focus the field and analysis efforts on those research questions that have not been addressed or for which adequate data have not yet been collected from the site. To achieve this goal, SEARCH developed a field strategy that utilized ground‐penetrating radar (GPR) and mechanical stripping to identify midden areas and features (trash pits, hearths, postmolds, etc.), the excavation of exposed features, a targeted investigation of midden and non‐midden areas using hand‐excavated test units to obtain stratigraphic data, the collection of faunal and botanical data from column samples, and the collection of radiocarbon samples from features and controlled contexts in test units. In the following sections, the research questions that guided the Bayou Park data recovery and analysis are described, along with the field methods used to recover the necessary data classes for addressing these questions. Laboratory methods related to the processing of artifacts and fauna also are discussed. Methods related to specific types of analyses are presented in individual chapters that discuss results of these analyses.
RESEARCH DESIGN In numerous reports and several published articles, Thomas and Campbell have laid out their argument for Elliott’s Point as representing a local manifestation of the larger Poverty Point phenomenon. More specifically, they consider the Bayou Park site to have been a particularly
Bayou Park, 8OK898
good example of a typical Elliott’s Point site. Chapter 2 provides a complete discussion of their argument as well as the relevant references. In Tables 3.1 and 3.2 we summarize some of their salient points, particularly those that can be evaluated with data from Bayou Park. While Thomas and Campbell present a persuasive case regarding the connection between Elliott’s Point and Poverty Point, we note that Bayou Park is one of the earliest recorded Late Archaic sites in the Choctawhatchee Bay area, with four radiocarbon dates indicating occupation during the initial phase of the Elliott’s Point Complex, or sometime between 4481‐ 3766 cal BP (2532 and 1817 cal BC). This span falls slightly before the earliest of dates recorded for the Poverty Point site in Louisiana (Connolly 2006), predating the onset of the significant mound building and exchange activities that would later come to define that regional center. This begs the question – if the Elliott’s Point Complex is defined in large part by its association with Poverty Point and the Poverty Point Exchange Network, yet one of its earliest sites pre‐ dates the fluorescence of Poverty Point, what then constitutes Elliott’s Point? How does the Elliott’s Point Complex relate to Poverty Point? And how does Bayou Park fit into existing interpretations of Elliott’s Point that were discussed in Chapter 2? To answer these questions, we first needed to develop or expand on existing baseline data regarding site chronology, site structure, subsistence and seasonality, technology, and site function. The specific questions related to these topics are outlined below.
Dating and Site Chronology Four C14 dates had been reported for the site prior to the 2011‐2012 excavation, but additional dates were desirable to refine the site’s chronology, particularly since we expected structural remains and features to be identified and excavated. Additional radiocarbon dates also would aid in both intra‐ and inter‐site analyses of subsistence practices, technology, and season of occupation, and would provide data that could be used to answer question regarding any changes in site use through time. Obtaining radiocarbon dates from well‐controlled contexts was thus a major goal of the project. To guard against the possibility of dating non‐cultural charcoal or roots, we planned to obtain samples for radiocarbon dating from charred wood and marine shells obtained from controlled stratigraphic contexts and features. Charred wood fragments can migrate vertically as a result of bioturbation (root action, burrowing animals and insects) and we intended to choose these samples carefully from sealed contexts that we were certain represented cultural events. In terms of shells, we intended to select robust shells, such as quahog clam (Mercenaria sp.), and if possible use only a single species for dating to limit between‐species variation in carbon uptake and fractionation (George Luer, personal communication, 2011).
Site Structure Morehead et al. (2011:24) have observed that “Although rich deposits have been encountered [at 8OK898], there is no observable pattern to the distribution. It is possible to excavate in one
70
Research Design and Methods
Table 3.1. Characteristic traits of the Elliott’s Point Complex. Traits References Elliott’s Point dates from approximately 2500‐600 BC Campbell et al. 2004:132; Campbell et al. 2009:17; Morehead et al. 2008:81 Represents a continuation of the well‐developed Late Thomas and Campbell 1993:528 Archaic adaptive pattern Represents a local manifestation of the Poverty Point Campbell et al. 2004:132‐133; Thomas and culture Campbell 1991, 1993:531; Involvement in a well‐developed Poverty Point trade Campbell et al. 2004:137; Thomas and network a “distinguishing attribute” Campbell 1991:108, 115, 1993:528, 531 Settlement pattern mirrors that found in the Lower Thomas and Campbell 1991:105‐106, 1993:528 Mississippi Valley Settlement is strongly coastal oriented Campbell et al. 2004:134; Campbell et al. 2009:20; Thomas and Campbell 1991:104 Oyster, quahog clam, and scallops are a diagnostic Thomas and Campbell 1993:541; Campbell et feature of Elliott’s Point sites al. 1998:77; Campbell et al. 2004:134; Campbell et al. 2009:20 Most Elliott’s Point sites represent camps clustered Thomas and Campbell 1993:528 around regional centers Regional centers arranged in oval pattern, smaller sites Thomas and Campbell 1991:105‐106, in semi‐circular pattern 1993:528‐531 Accretional mounds are characteristic, e.g., Meig’s Campbell et al. 2009:18 Pasture, Buck Bayou Lithic workshops separated from mound or living areas Thomas and Campbell 1991:106‐107, 1993:531‐532 Regional centers served as loci of community feasts, Thomas and Campbell 1993:531‐532 redistribution centers for exotic, non‐local goods Earliest evidence of Elliott’s Point Complex consists of Thomas and Campbell 1993:533, 541 “crude, amorphous baked‐clay objects” in association with Late Archaic projectile points Well‐formed baked‐clay objects, particularly spheroids, Thomas and Campbell 1991:108, 1993:533 are “typical” of “classic” Elliott’s Point Complex Baked‐clay objects (Elliott’s Point Objects) “clearly” Thomas and Campbell 1993:532 linked with Poverty Point Microlith industry similar to Poverty Point (Jaketown Campbell et al. 2004: 133; Thomas and perforators) Campbell 1991:108, 1993:533 Stemmed projectile points are “directly comparable” to Campbell et al. 2004:133; Thomas and projectile points found at Poverty Point, Claiborne, and Campbell 1993:535 Cedarland A particular style of hafted biface, referred to as a Destin Thomas and Campbell 1991:110‐111, 1993:536 point, is characteristic of Elliott’s Point No well‐developed lapidary industry Thomas and Campbell 1991:108, 1993:534 A “moderately active” bone and shell industry” is Thomas and Campbell 1991:108, 1993:535 characteristic Bone and shell beads may have been manufactured for Thomas and Campbell 1993:535 export in exchange for non‐local goods Fiber‐tempered pottery introduced late in the Elliott’s Campbell et al. 2004:139; Thomas and Point chronology as an addition to well‐established Campbell 1993:538‐539 complex.
71
Bayou Park, 8OK898
Table 3.2. Characteristics of Bayou Park site. Characteristics References Contains all the elements necessary for Elliott’s Point: Campbell et al.:1998:77 EPOs, Jaketown perforators, exotic items, “unusual” shell composition (oyster, clam, scallop) Portion of excavated midden area was the focus of food Campbell et al. 1998:78‐79 preparation activities Possible seasonal (fall, winter) occupation (hickory nuts) Campbell et al.:1998:86 Possible late summer through fall occupation (scallops, Morehead et al. 2011:114 hickory nuts) “At least semi‐permanent” habitation, but year round Meyer et al. 2002:64; Morehead et al. occupation cannot be ruled out 2011:114 Northwestern area investigated in 2002 posited as a Meyer et al. 2002:72 special activity area due to pieces esquillees, a rod, and a double‐backed flake; possibly bead manufacturing. Compared with 8WL36 Feature 4 in southeastern area interpreted as part of a Meyer et al. 2002:64 feasting event Lithic technology oriented toward biface production and Morehead et al. 2008:79 maintenance Possible earlier introduction and subdivision of EPC based Campbell et al. 2004:132 on C14 dates from 8OK898 Possible living surface between 40 and 70 cm in non‐ Morehead and Aubuchon 2008:71‐72 midden area No observable pattern to the distribution of deposits Morehead et al. 2011:24
area and uncover dense shell midden and features, and then place excavation units less than 10 m away and find no evidence of Elliott’s Point remains whatsoever…. Similar situations have been encountered at a series of Elliott’s Point sites along Rocky Bayou and Rocky Creek where multiple investigations failed to reconstruct intra‐site residential versus activity patterning….” Although features (including a postmold) have been identified during previous investigations, there has been no large‐scale exposure of subsurface features sufficient to identify the structural layout of the site. Nor has there been sufficient exposure of midden deposits to determine the true nature and extent of these deposits. If overlapping features are present, this would suggest reuse and (perhaps) reoccupation of areas over time. PTA’s 1998 block excavation of a small portion of the site’s midden indicated that shell and non‐shell areas are present. Some of the questions that we wished to answer concerned the nature of these shell and non‐shell deposits. Do the shell areas represent discrete deposits of one or a few episodes of refuse disposal? Or are they lenses representing more continuous disposal over a lengthy period of time? Are the non‐shell areas devoid of fauna or do they contain vertebrate remains? How large is the midden? If structures could be identified through the exposure and arrangement of postmolds or postholes, another set of questions arise. It might be possible to estimate structure size and thus infer population density at the site. The size of structures, and the size and numbers of posts associated with them, can provide information on the degree of architectural investment by site occupants which can then be used to infer intensity of occupation. How do these data compare with what is observed locally and regionally? What 72
Research Design and Methods
might the degree of occupation intensity mean in terms of interaction, association, or environmental adaptation? Several spatially distinct areas of artifact, midden, and feature concentration within 8OK898 suggest that there may be functional or temporal differentiation across the site in terms of activities performed. For example, lithic reduction areas are separated from living areas at some Elliott’s Point sites. Is there evidence for a similar separation of activity areas at Bayou Park? Additional data collection could determine if these spatially distinct areas represent functionally different activity areas within a single site complex, or separate campsite locations that were utilized at different points in time. Finally, we wished to determine if any other evidence of organized site structure exists. For example, does the midden conform to the arcuate structure identified at other Late Archaic sites in the region? Thomas and Campbell (1991:105) suggest that this is a trait shared with Poverty Point sites in the Lower Mississippi Valley, while Russo (2004) suggests that arcuate and circular shell “rings” are the loci of communal feasting events. If a so‐called plaza exists, what were the activities that were conducted within this social space? What does the presence or absence of formal site structure suggest about the earliest Late Archaic expressions on the Choctawhatchee Bay?
Subsistence A wide range of subsistence remains have been identified at 8OK898, including shellfish, fish, deer, turkey, turtle, squirrel, and bird. However, these previous excavations utilized a 1/4‐in (6.4‐mm) mesh screen for recovery. Numerous studies have demonstrated the necessity of using fine‐mesh screens to obtain a more representative sample of vertebrate faunal remains, particularly those associated with fish (e.g., Quitmyer 2004; Wing and Quitmyer 1985). That only four species of fish have been reported for 8OK898 (S. L. Scott 2002), underscores this fact since coastal sites typically contain many more fish species. In addition, quahog, scallop, and oyster shell have been referred to as a “hallmark” of the Elliott’s Point Complex (Campbell et al. 1998:77), but no detailed faunal analysis of an Elliott’s Point site has ever been conducted. Moreover, the only botanical remains reported from the site include a few hickory nut shells, five unidentified plant specimens, and one unidentified seed from a bell‐shaped feature (Campbell et al. 198:75‐76; Roberts 2002). Accordingly, a primary concern of the data recovery project was the recovery of bulk samples for detailed faunal and botanical analysis. Not only were we interested in characterizing the diet of the site’s inhabitants, but we wanted to compare these data to similar data from other Late Archaic sites in the Choctawhatchee Bay region.
Seasonality and Settlement Patterns Bayou Park has been tentatively interpreted as a seasonal (fall‐winter) habitation based on the recovery of hickory nut shells (Campbell et al. 1998:76, 86; Meyer et al. 2002:40). Morehead et al. (2011:114) expanded this to include late summer (July‐August) based on the presumed seasonal availability of scallops in the bay during those months. The seasonal occupation of the site needed to be confirmed with additional data, specifically the analysis of clam shell growth rings and detailed botanical analysis. Was the site truly occupied during only one or two
73
Bayou Park, 8OK898
seasons, or could it have been a permanent habitation? If the site was occupied seasonally, was there a consistency over time in the season of use, or could it have been occupied during multiple seasons? What does seasonality or multi‐seasonality suggest about early Late Archaic society? How does Bayou Park compare with the diversity of subsistence and settlement strategies observed both along the Gulf coast and regionally? Seasonal occupation implies settlement mobility and the most likely scenario is movement between the coast and the interior. Because there are no significant toolstone resources along the northwest Florida coast, movement to and from interior locales should have resulted in the deposition at Bayou Park of lithic tools and manufacturing debris of raw materials obtained from interior source areas (Lloyd et al. 1983; Upchurch et al. 1982). Determining if these materials were obtained directly, implying movement of people, or indirectly via exchange, is difficult; however, the size and form of lithic and materials might provide clues as to whether they were procured directly or indirectly via exchange (e.g., Austin 1997:150‐155, 158‐160).
Site Function Site function does not appear to have been addressed in previous studies beyond the designation of the site as a seasonal habitation. A formal analysis of site function taking into account seasonality data, number and types of features, evidence of structures and their sizes, the presence of interior or exterior hearths, and their locations is necessary to adequately address this issue. Changes in site function through time may be determined depending on the number of features and radiocarbon dating of various features. An important aspect of site function is establishing the intensity of site use. Differential patterning in site intensity is a reflection of different types of types of settlements, for example a nucleated village versus a dispersed campsite. This is based on the assumption that the amount of material remains deposited at a site is a function of the span of time that is spent at a specific location during the annual cycle and the cumulative time over which the site has been used (Lightfoot 1984:89). By plotting these variables on X and Y axes, it is possible to identify four idealized types of archaeological sites based on intensity of use (Figure 3.1). To operationalize this model, we planned to use a method devised by Gallivan (2002). Using archaeological data that had been shown to be highly correlated with one another, Gallivan developed two indices that could be plotted according to Lightfoot’s heuristic model. Residential stability is determined using the following measures: feature richness, lithic evenness, mean house floor area, mean post diameter, mean number of structural wall posts, mean number of interior features, and mean pit volume. The average of the resulting values is the “residential stability index” (Gallivan 2002:542). A similar index attempts to identify the use duration of a site by averaging the densities of features, postmolds, burials, and artifacts, in addition to the ratio of burials to houses (Gallivan 2002:543). The resulting measures are standardized so that values range between 0 and 1 and these are then plotted on a bivariate graph, allowing placement of individual sites within the four idealized categories shown in Figure 3.1. 74
Research Design and Methods
1
Multiple, year-long occupations
Use Duration
Multiple, brief occupations
Single, brief occupations
Single, year-long occupations
0 0
1
Residential Stability
Figure 3.1. Model of occupational intensity (after Lightfoot 1984).
Except for burials, Bayou Park contains all of the data classes used by Gallivan in his study and has the added benefit of containing faunal remains. We hoped to be able to develop similar indices with these data and compare the results to more conventional analyses of assemblage diversity and organizational structure to arrive at an interpretation of site function.
Technology Two major classes of artifacts have been identified at Bayou Park: baked‐clay objects (aka Elliott’s Point Objects) and stone tools and manufacturing debris. Neither has received detailed analysis, although subjective comparisons between sites have been made in several reports.
Baked‐Clay Objects Baked‐clay objects are one of the distinguishing characteristics of Elliott’s Point Complex sites and one of the chief reasons for positing a relationship with Poverty Point; but there are many unanswered questions about their origins and use. What types of baked‐clay objects are present at Bayou Park? What might the variety (or lack thereof) of baked‐clay objects suggest about the development of the technology or the utilization of specific cooking technologies during the Late Archaic? Are there diachronic or synchronic patterns of baked‐clay object types present? What might the presence or absence of these patterns mean in terms of site organization or site utilization through time? How do these data compare to what is observed both locally and regionally throughout the Late Archaic Period? How do they compare to other sites with baked‐clay objects, including Poverty Point? Were the better formed, deco, rated objects acquired via trade, or were all baked‐clay objects manufactured locally? We hoped to be able to address some of these questions through detailed analysis of paste and temper
75
Bayou Park, 8OK898
characteristics, between site comparisons of assemblage composition, and the use of instrumental analyses to compare selected objects with local clay sources. Microliths Microliths are part of the material culture package that is used to infer a relationship between Elliott’s Points sites and Poverty Point‐related sites. While microlithic stone tools are relatively common at Elliott’s Point sites, they are not abundant at Bayou Park. Is this due to sampling bias or is it a true reflection of the stone technology of the site’s inhabitants? If microliths are indeed rare at Bayou Park, but common at other contemporaneous sites, what could be the reason? There has been no detailed analysis of lithic technology and raw material acquisition strategies, nor any functional analysis of stone tools. What were Elliott’s Point microliths used for and how closely do they resemble the Jaketown Perforators and other microlithic tools found at Poverty Point‐related sites elsewhere? Projectile Points Elliott’s Point sites appear to contain a wide variety of stemmed bifaces, some of which are claimed to resemble types found at Poverty Point and other related sites (Morehead et al. 2011:111‐114; Thomas and Campbell 1991:108‐111, 1993:535‐536). This includes a unique type reportedly identified by Clarence Webb and referred to locally as a Destin point (Thomas and Campbell 1991:110‐111). Morehead et al. (2011:112) have noted the basic similarity in the haft areas of the various stemmed points identified at Bayou Park; that is, they tend to have sloping shoulders, a straight to slightly contracted stem with concave edges, and an essentially straight base. This distinguishes them from the Destin type, which tends to have a concave stem base. Is the Destin point a true diagnostic that can be consistently separated from other Late Archaic stemmed points? Or is it simply a “variation on a theme”? Alternatively, what do the straight‐based stemmed points at Bayou Park represent? Do they reflect a different social group or are they simply Destin variants? Finally, how does the overall lithic assemblage compare with other Late Archaic sites in the panhandle and what might similarities or differences in these assemblages suggest about settlement and interaction?
Poverty Point Relationships As previously discussed, the Elliott’s Point complex is believed to be “a localized expression of the Poverty Point culture” (Thomas and Campbell 1991). Similarities in settlement patterns and artifact assemblages, along with the presence of exotic artifacts at some sites, form the basis for this interpretation. The review of previous research in Chapter 2 demonstrates that similarities between Elliott’s Point and Poverty Point sites do exist; however, the interpretation of these similarities and what they represent has been contested, most notably by Nancy White (2003b, 2004), who suggests that similarities in material culture may simply be the result of similar site functions and subsistence practices by coastal‐oriented groups. While it is unlikely that the debate over Elliott’s Point and Poverty Point relationships can be resolved with data from a single site, there are some questions that can profitably be investigated at Bayou Park. Certainly material culture comparisons were a major goal of the project. Do the baked‐clay objects from Bayou Park closely resemble those from Poverty Point‐related sites, or do they, as White has suggested, represent the use of a similar technology to address a similar problem, 76
Research Design and Methods
that is, the absence of stone suitable for dry or wet heat cooking? Does the microlith assemblage at Bayou Park and other Elliott’s Point sites truly resemble the microlith assemblages from Poverty Point, Jaketown, and other related sites, or is it a local manifestation of a more widely used technology? Is the settlement pattern of small habitation sites and campsites surrounding a regional center a valid interpretation of the archaeological data? Is the wide‐spread appearance of arcuate shell middens a cultural or social construct (Russo 2004), or is it simply an artifact of landscape and site formation processes (Marquardt 2010a, 2010b)? Is there any evidence of trade at Bayou Park? What types of trade items are represented? What might the presence or absence of exotic items suggest about this site relative to other early Choctawhatchee Bay Late Archaic sites in terms of significance and interaction? What are the overall implications of the data recorded from 8OK898 regarding site utilization, structure, and seasonality on a local and regional scale? How does the development of the Late Archaic populations around the Choctawhatchee Bay compare to the greater development of Late Archaic traditions and what might this means in terms of long‐ distance interaction and association or identity?
Site Formation A final area of interest is the archaeological deposits themselves. In particular, we were interested in examining the efficacy of excavating only where artifacts or midden deposits are abundant. Most investigations of archaeological sites in Florida follow this approach under the assumption that areas of high artifact density represent “activity loci” while areas of low artifact density have tended to be overlooked (Austin 2002; Mathis 1993). Since much of the archaeological record is the result of human discard practices, the possibility that areas of low artifact density might contain evidence of behaviors other than those related to ceramic and stone tool use and discard is a serious issue from both the research and cultural resource management perspectives. Because we intended to expose large areas using heavy equipment to look for features, this seemed to offer an opportunity to test this assumption by comparing the spatial distributions of archaeological features with the spatial distributions of midden deposits or other concentrations of artifacts that had been identified previously through broad‐ scale testing. If the correlation between features and middens or artifact concentrations is positive, then our strategies of investigating most intensively those areas where cultural materials are abundant are justified. However, if the correlation is negative, then greater efforts should be made to investigate low‐density areas during testing and excavation projects.
FIELD METHODS Prior to fieldwork, the project area was divided into three arbitrary areas – Operations A, B, and C – as shown on Figure 3.2. These areas served to delineate portions of the site for the purposes of obtaining AF‐103 Civil Engineering Work Clearance for buried utilities. An extension of the site to the east along Postl Lake (labeled B‐extension on Figure 3.2) was not included in the AF‐103 permit because no excavation was planned in this low‐density part of the site, although ground‐penetrating radar work was conducted there.
77
Figure 3.2. GPR grids within Operations A, B, and C.
Bayou Park, 8OK898
78
Research Design and Methods
Before discussing the field methods, we should explain the use of directions when referring to the orientation of survey areas, mechanical strip areas, and test units. The SEARCH grid system established for the site is based on magnetic north. While most test units also are oriented to magnetic north, the ground‐penetrating survey areas and the mechanical strip areas are oriented parallel to existing buildings and sidewalks, which are oriented parallel to existing roads, or at angle of approximately 35° east of north. To avoid confusion and facilitate communication in the field, we established an operational “north” for the survey and strip areas which corresponded to northeast on our grid system. Throughout the report, when referring to these operational directions, we will place them in quotes to distinguish them from magnetic north.
Ground‐Penetrating Radar Survey The first task of the data recovery effort was a ground‐penetrating radar (GPR) survey of the site. The survey was conducted in an effort to identify subsurface midden deposits and possible features, modern utility lines, and other subsurface anomalies. GPR uses electromagnetic waves generated from an antenna as the energy source from which measurements are collected and interpreted. The radar energy that is propagated into the ground is reflected off buried objects, features, and changes in soil types (Conyers 2004). The lapse in travel time and varying amplitude of the electromagnetic waves collected along transects are interpolated to create a two‐dimensional image of the subsurface that are used to identify soil disturbances that could signal the presence of archaeological features, buried objects, grave shafts, and changes in sediment type. Because of the presence of buildings, sidewalks, and other obstacles, Operations A, B, and C were subdivided into sections using convenient landmarks (see Figure 3.2). To ensure accurate georeferencing of located anomalies, a datum stake was placed in a selected corner of each grid and the location of the stake was determined using a handheld Trimble XH Series 6000 GPS. Once the site grid system was established (see below), each grid datum was recorded in relation to the site datum (1000N, 1000E) with a Sokia Model 50X total station (Table 3.3). Geophysical data were collected using a GSSI/SIR‐3000 Integrated Radar Control Unit, mounted on an UtilityScan wheel‐cart with a 400‐MHz shielded antenna (Figure 3.3). The wheel cart containing the GPR unit was pushed along transects spaced 50 cm apart, and transect start and stop locations were documented on field data sheets to cross‐reference accuracy during post‐ processing. The GPR data were processed and analyzed with GPR‐SLICE (Version 7.0) imaging software, and GSSI’s RADAN (Version 7) GPR processing software.
Grid System and Site Mapping A previous grid system was established in 2002 by PTA (Meyer et al. 2002). Several permanent datum points also were established and elevations in meters above mean sea level (amsl) were obtained for these points. Additional datum points were established during a 2009 testing project by PTA (Campbell et al. 2009). Unfortunately, except for the building corners and a manhole cover, none of these datum points could be relocated. This required that the original grid datum be reestablished from existing field notes. According to the 2002 PTA report and
79
Bayou Park, 8OK898 Table 3.3. GPR grid sizes and location data.
GPR Grid #
1 2 3 4 5 6 7 8 9 10 11 12 15 16 17 18 19 24 41 42 43 Subtotal 25 26 27 28 29 30 31 32 33 34 35 36 37 Subtotal 14 20 21 22 23 38 39 40 44 Subtotal 13 Total
80
A A A A A A A A A A A A A A A A A A A A A
Northing, Easting 995N, 966E 979N, 990E 960N, 1020E 967N, 1043E 1012N, 1022E 1050N, 1047E 1004N, 1014E 996N, 1046E 1010N, 1083E 1036N, 985E 1049N, 1008E 1030N, 1088E 945N, 1042E 1062N, 1001E 1079N, 1013E 1110N, 1032E 1101N, 1042E 1056N, 1054E 1124N, 1102E 1097N, 1079E 1157N, 1086E
Grid Datum Corners UTMs (Northing, Easting) 3371629N, 549017E 3371613N, 549041E 3371594N, 549070E 3371601N, 549095E 3371646N, 549073E 3371682N, 549098E 3371638N, 549065E 3371629N, 549098E 3371642N, 549134E 3371670N, 549036E 3371683N, 549058E 3371700N, 549165E 3371577N, 549093E 3376940N, 549052E 3371711N, 549063E 3371741N, 549082E 3371733N, 549093E 3371690N, 549105E 3371756N, 549153E 3371729N, 549130E 3371789N, 549137E
B B B B B B B‐EXT B‐EXT B‐EXT B‐EXT B‐EXT B‐EXT B‐EXT
910N, 1204E 919N, 1249E 964N, 1268E 944N, 1283E 946N, 1260E 986N, 1269E 948N, 1328E 985N, 1351E 1001N, 1391E 1030N, 1446E 1038N, 1445E 1002N, 1298E 987N, 1291E
3371545N, 549256E 3371554N, 549301E 3371596N, 549319E 3371576N, 549332E 3371578N, 549311E 3371618N, 549320E 3371580N, 549379E 3371617N, 549402E 3371633N, 549442E 3371662N, 549497E 3371670N, 549496E 3371636N, 549349E 3371619N, 549342E
C C C C C C C C C
987N, 1120E 951N, 1110E 939N, 1143E 987N, 1162E 997N, 1156E 1063N, 1194E 1049N, 1181E 1056N, 1177E 1050 N, 1216 E
3371621N, 549171E 3371583N, 549161E 3371573N, 549194E 3371620N, 549213E 3371629N, 549207E 3371695N, 549245E 3371681N, 549232E 3371688N, 549228E 3371682N, 549268E
970N, 1084E
3371605N, 549136E
Operation
A&C
Size (Acres) 0.301 0.099 0.094 0.012 0.315 0.030 0.089 0.334 0.062 0.038 0.010 0.473 0.019 0.040 0.158 0.059 0.077 0.037 0.040 0.067 0.053 2.429 0.055 0.036 0.086 0.047 0.030 0.033 0.086 0.100 0.052 0.098 0.040 0.069 0.034 0.900 0.225 0.133 0.145 0.064 0.033 0.064 0.015 0.016 0.010 0.752 0.202 4.081
Research Design and Methods
Figure 3.3. GSSI GPR unit used for data collection at 8OK898.
field maps on file at Eglin AFB, the 1000N, 1000E point was located about six meters to the northeast of a concrete sewer marker in the northwest corner of the prison camp. This location corresponded roughly with the common corner of TPs 7, 11, and 14 in PTA’s 1998 block excavation. The concrete marker was recorded as 996N, 964E on field maps for the 2002 project, with angles and distances recorded to the corners of Buildings 588 and 591. In contrast, the 2009 report records the concrete marker as being located at 999.1N, 964.4E. It does not appear that a permanent hard point, such as steel rebar, was used to mark the 1000N, 1000E datum originally, and conversations with Jan Campbell of PTA and Joe Meyer, currently with Eglin AFB, failed to resolve the discrepancy or positively identify either the concrete sewer marker or the exact location of 1000N, 1000E. Therefore, it was necessary to use the total station and the locational data recorded on the 2002 field map to relocate and reestablish both of these points. Using this method it was determined that a square concrete sewer marker located near the intersection of Inverness Road and Flagler Road (Figure 3.4), was the probable concrete sewer marker referenced in both reports. It should be noted, that the angles and distances on PTA’s field map do not match exactly the location of this concrete marker. In addition, as there was no way to know for certain which of the two coordinates for this location is correct, we assumed the original coordinates of 996N, 994E for the concrete sewer marker and using this as a starting point, PTA’s 1000N, 1000E datum was reestablished. Subsequent stripping, which revealed the outline of the 1998 excavation block, indicated that the reestablished datum location is approximately ~ 1.24 m southeast of the original datum (Figure 3.5). A two‐foot long piece of steel rebar was set at this new location as a permanent marker
81
Bayou Park, 8OK898
Figure 3.4. View to the northeast of Operation A showing locations of the concrete sewer marker, PTA’s 1998 block excavation, and SEARCH’s grid datum. (UTM E549051 N3371632, Zone 16, WGS84). A second permanent datum stake (two‐foot long
steel rebar) also was set at 1000N, 1200E (UTM E550251 N3372632). The elevation of the concrete sewer marker is recorded in the 2009 report as 3.30 m amsl and the elevation of a second datum (a manhole cover located north of Building 633, which was identified in the field) is recorded as 3.88 m amsl. Using these two elevations and the reestablished 1000N, 1000E datum location, SEARCH established the grid locations and elevations of 60 hard points across the site including all of the lamp poles, several building corners, and miscellaneous electrical boxes and fire hydrants. These included several of the building corners that were used as hard points by PTA. These data are provided in Appendix 3.1. Our coordinates and elevations for their hard points were generally within tolerable limits (± 1 m) of the original documented data such that we felt confident that PTA’s grid system and datum plane were reestablished as accurately as possible. The 60 hard points were used throughout the project to establish locations and elevations for strip areas, test excavation units, and most importantly, features exposed during mechanical stripping. In addition, all of the GPR datum stakes were tied into the grid system and datum plane. Finally, the locations of all hard points, including the 1000N, 1000E and 1000N, 1200E datum stakes, were recorded with the Trimble GPS.
82
Research Design and Methods
TP60
Figure 3.5. Relative locations of SEARCH and PTA 1000n, 1000E datum points within Operation A.
Aligning PTA Excavations with SEARCH Grid When available, we used the grid coordinates recorded in PTA field notes or maps included with PTA reports to determine the locations of previously excavated shovel tests and test units. Unfortunately, PTA excavations conducted prior to 1998, and some excavated after this date, have no grid coordinates associated with them and it was necessary to estimate their locations using maps available in PTA reports and from field notes. Moreover, the compass and tape method used to record many of the shovel test and test unit locations originally (Joe Meyer, personal communication, 2012) are less accurate than the total station resulting in random measurement errors. In general, the procedure that was used to reconcile the two systems involved using the UTM coordinates for the 60 SEARCH hard points that were obtained with the Trimble GPS, then plotting these on a georeferenced aerial photograph of the site in GIS (Zone
83
Bayou Park, 8OK898
16, WGS84). Using the SEARCH 1000N, 1000E datum as a reference, all of the PTA shovel tests and test pits with coordinates also were converted to UTMs and plotted on the same aerial. All of the PTA tests without coordinates were manually added to the aerial based on their relation to one another and known landmarks (e.g., building corners, sidewalks, etc.). Their UTM coordinates were then calculated in GIS and these were then converted mathematically to the SEARCH grid system. Because the two grid systems do not align perfectly, we knew that adjustments would need to be made so that previous excavations could be matched as closely as possible with the SEARCH grid system. When comparing the locations of PTA tests on the GIS map to PTA report or field maps, it was clear that some of these tests were not plotted correctly on the GIS. However, we were fortunate in being able to identify PTA’s 1998 excavation block during stripping as well as their TPs 19 and 20, which enabled us to georectify some locations. We have discussed the excavation block above, but will describe the two test unit locations here because they illustrate the difficulty in matching up the two grid systems and provide justification for our final decisions regarding grid alignment. In Strip Area K, a 1‐x‐1‐m square area of mottled soil was identified which we believe to be PTA’s TP19. Two smaller, square stains also were identified in this strip area and were identified as PTA’s STs 167 and 168. These three excavations were conducted during a 2002 survey of a sprinkler and communication trench (Meyer et al. 2002). The grid location of TP19 recorded in PTA field notes is 1026N, 1041E; on the SEARCH grid the coordinates are 1024.55N, 1040.55E or a difference of ‐1.45N, ‐.45E. Similarly, the grid coordinates for ST 167 are reported in field notes as 1009N, 1035E and those for ST168 as 1027N, 1048E. Using the SEARCH grid, the coordinates for these shovel tests become: ST167, 1013.31N, 1032.92E; ST168, 1013.20N, 1028.10E. The differences from the recorded coordinates are much greater for the shovel tests than for TP19 and this is likely due to the methods used to record their locations originally in the field; i.e., compass and tape. Considering the close agreement between the two coordinate systems for TP19, as well as the fact that no other large, square excavation is recorded as being close to the 1‐x‐1‐m stain revealed by stripping, and assuming the relational accuracy of the shovel tests and test units on the PTA maps, we feel confident in stating that the two smaller square stains are indeed STs 167 and 168. Using these three units as reference points, it was then possible to visually shift other shovel test locations along the 2002 sprinkler and communication trench alignment to more closely match their locations in relation to Buildings 591, 626, and 632 as shown on PTA maps. PTA’s TP20 was more difficult to locate. The coordinates for this unit, which actually consisted of two overlapping 2‐x‐2‐m units excavated to expose a feature, are recorded in field notes as 927N, 1252.7E. Using the total station, we reestablished this location and used the hydraulic excavator to strip the area in hopes of seeing the characteristic straight‐sided, mottled soil stain indicative of a back‐filled excavation unit. None was identified. We then used measurements obtained from PTA’s field maps to measure the TP’s location in relation to Building 636. This area was then stripped and the southern edge of TP20 was identified. The SEARCH coordinates for the unit are 933.63N, 1238.30E. Using this location as a reference point, we then adjusted 84
Research Design and Methods
the shovel tests excavated along the sprinkler line to more closely match their relationship to TP20 and existing buildings. Final manual adjustments to other PTA tests were made as necessary to match the locations shown on PTA maps as closely as possible. These were converted to the SEARCH grid system using the UTM/grid conversion method described above. The new coordinates were then used for all subsequent mapping and distributional analysis. This lengthy discussion has been necessary to explain the differences in grid coordinates between the two grid systems. Certainly one potentially complicating problem is the reporting of two different sets of coordinates for the original concrete sewer marker, from which all other coordinates (including both the PTA and SEARCH 1000N, 1000E data points) were measured. Our decision to use the original set of coordinates rather than the second, revised set of coordinates can be questioned, but at the time we had no good reason to choose one set of coordinates over the other. We chose the original set because we felt that these gave us the best opportunity for relocating PTA’s original 1000N, 1000E datum. In hindsight, using the second set of coordinates might have lowered the margin of error between the two systems, particularly for later excavations. On the other hand, the fact that we could determine no constant mathematical correction regardless of which system was used suggests that the location errors were random and adjustments would have needed to be made regardless of which system was used. For those who wish to use them, we have included the original and revised PTA coordinates along with the new SEARCH coordinates in Appendix 3.2.
Mechanical Stripping The large number of utility lines that crisscross the site, as well as the presence of numerous structures and sidewalks, made site‐wide stripping with heavy equipment impractical. In addition, as a condition of the AF‐103 process, only hand‐digging was allowed within three feet of any buried cable lines. Consequently, mechanical stripping of selected areas was performed instead. Based on the previous work conducted at the site, the areas that seem to offer the best potential for stripping included the area in and around Buildings 588 and 591 (both of which had been recently demolished), and the open field near PTA’s 1998 block excavation both located within Operation A. The goal was to continue to expose, excavate, and analyze the midden and feature deposits in this portion of the site. Operation A also contained midden deposits with abundant faunal remains which we planned to collect for analysis with column samples. Operation B, in the southeastern portion of the site nearest Postl Lake, included the location where PTA excavated a large pit feature and recovered a large number of finished tools (TP20). Stripping of selected areas here was planned in order to determine if other features are present and if so to gather data and radiocarbon dates for comparison with the midden in Operation A. Operation C included the central courtyard and several buildings. Limited stripping was planned in this area between the existing sidewalks where few artifacts had been found previously. Finally, the location of a sandstone feature identified during monitoring to the north of Building 635 in 2003 also was planned for stripping. 85
Bayou Park, 8OK898
Figure 3.6. View to the “north” of Strip Area K with hydraulic excavator. Note midden deposits on the left.
In all, mechanical stripping was performed in 30 separate areas and was accomplished using a Caterpillar tracked hydraulic excavator with a 36‐inch ditch clean‐out (toothless) bucket to remove overburden and expose subsurface features (postmolds, pits, etc.) (Figure 3.6). The equipment was selected for its size, tracked wheels, and maneuverability to minimize any potential damage to exposed archaeological deposits, existing sidewalks, and lawns. Individual strip areas were labeled by letters (e.g., Strip Area A, B, C, etc.) and these are shown on Figures 3.7‐3.9. The placement and size of the strip areas were determined by several factors, including previous survey and test excavation results, the GPR results, and the need to avoid various obstacles (buildings, sidewalks, buried utilities, and large trees). Nineteen were located in Operation A, eight in Operation B, and three in Operation C. Table 3.4 provides data on the size and volume of each of the 30 Strip Areas. All mechanical stripping was monitored by an archaeologist and was conducted according to the following procedure. The hydraulic excavator first removed fill and overburden where present. The boundaries and surface elevations of each strip area then were recorded with the total station in relation to the site grid system and datum plane. Excavation with the hydraulic excavator then proceeded in 10‐to‐15‐cm increments in order to expose large areas in plan view (see Figure 3.6). At the completion of each 10‐to‐15‐cm “level,” exposed surfaces were shovel‐shaved and any midden deposits, features, or diagnostic artifacts were photographed and mapped, and their locations and elevations recorded with the total station. Selected features also were excavated at this time (see Excavation Methods below). After all mapping and hand excavation were completed, stripping resumed and was carried out in a similar
86
Research Design and Methods
Figure 3.7. Strip areas and test pit locations in Operation A.
87
Bayou Park, 8OK898
Figure 3.8. Strip areas and test pits locations in Operations A and C.
fashion until no additional midden deposits or features were encountered. The termination depths varied, but usually were no less than 1.25 m below ground surface, with several trenches excavated to approximately 1.5 m below ground surface. At least one wall in each strip area was cleaned and photographed, and profile drawings were created to document the stratification of the entire site from its northwestern edge, near building 591, to its southeastern terminus near Postl Lake. In addition, a perpendicular profile from several trenches located in Operation A also were photographed and drawn in order to document stratification through the midden area from southeast to northeast. Stratum designations for all profiles followed existing nomenclature established in previous work by PTA (see Chapter 4). Bottom elevations were obtained with the total station for each strip area prior to backfilling.
Excavation Methods A total of 30 test excavation units of various sizes and eight 50‐cm‐in‐diameter shovel tests were excavated at 8OK898 (see Figures 3.7‐3.9). Continuing the practice established by PTA, each unit and shovel test received a unique identifying number as well as a grid coordinate
88
Research Design and Methods
Figure 3.9. Strip areas and test pit locations in Operation B.
89
Bayou Park, 8OK898
Table 3.4. Size and volume data for mechanical strip areas. Strip Area A B C D E F G H I Ja Jb K L M N O Pa Pb Q R S T U V Wa Wb X Y Z AA Totals
Operation A A A A A A A A A A A A A B B A B B A B B B B A C C C A A A
Perimeter (m) 18.86 58.53 39.83 56.48 71.86 39.51 19.82 36.95 39.25 92.37 77.84 96.59 68.81 40.02 52.02 41.53 20.01 127.96 72.86 48.71 18.93 26.56 34.09 57.28 20.91 18.10 44.64 88.46 31.61 27.43
Area (m2) 21.82 61.32 93.62 150.30 190.66 75.71 24.17 72.90 75.08 303.94 249.36 291.02 182.97 56.73 131.67 71.56 16.82 137.42 223.59 77.05 20.71 32.26 51.83 162.38 20.76 19.00 92.93 159.12 46.49 46.78 3113.19
Volume (m³) 32.74 91.98 140.42 225.45 285.98 113.57 36.25 109.35 112.62 455.91 374.04 436.53 274.45 85.10 197.50 107.34 25.24 206.13 335.39 115.57 31.07 48.39 77.75 243.57 31.14 28.50 139.39 238.69 69.73 70.16 4669.79
designation. The southwest corner was used as the identifying pair of coordinates for the test unit while the center points were used for the shovel tests. The total volume of excavated sediments from these units was 76.21 cubic meters (Table 3.5). Shovel tests were judgmentally excavated within Building 588 (n=3), Building 591 (n=3), and just outside and to the northwest of Building 591 in order to determine the presence or absence of midden within and adjacent to these demolished buildings. The shovel tests were excavated to 100 cmbs unless dense midden was encountered, in which case they were
90
Research Design and Methods
Table 3.5. Grid coordinates and elevation data for SEARCH shovel tests and test pits. Grid Coordinatesa, b Elevation Volume Type Number Size Operation (m)b (m3) N E ST 257 Void ST 258 50 cm dia. 977.00 1038.10 3.04 0.20 A ST 259 50 cm dia. 983.81 1028.80 3.17 0.20 A ST 260 50 cm dia. 995.88 1012.80 3.45 0.06 A ST 261 50 cm dia. 1033.64 1013.29 3.68 0.20 A ST 262 50 cm dia. 1039.90 1018.34 3.61 0.09 A ST 263 50 cm dia. 1053.60 1025.24 3.71 0.20 A ST 264 50 cm dia. A 1040.94 1010.10 3.76 0.20 ST 265 50 cm dia. A 1043.67 1012.00 3.97 0.20 TP 58 1‐x‐2 3.61 2.00 A 976.98 1035.87 TP 59 2‐x‐2 3.19 4.00 A 984.46 1028.40 TP 60 1‐x‐2 3.42 2.00 A 994.99 1010.61 TP 61 1‐x‐2 3.22 1.92 A 987.03 1021.71 TP 62 1‐x‐2 3.71 1.94 A 1052.58 1025.5 TP 63 1‐x‐2 3.61 2.00 A 1037.29 1018.16 c TP 65 1‐x‐1 1010.39 1002.83 2.90 0.70 A TP 66 1‐x‐1 1011.13 1003.48 2.99c 0.69 A TP 67 1‐x‐1 1012.87 1004.18 3.02c 0.57 A TP 70 1‐x‐1 1009.69 1003.55 2.50c 0.32 A c TP 71 1‐x‐1 1010.45 1004.22 2.29 0.29 A TP 74 1‐x‐1 1008.32 1003.61 2.93c 0.67 A TP 75 1‐x‐1 1009.01 1004.28 2.89c 0.96 A TP 76 1‐x‐1 1009.78 1004.97 2.83c 1.02 A c TP 79 1‐x‐1 1007.64 1004.31 3.11 0.72 A TP 80 1‐x‐1 1008.36 1005.00 3.08c 0.91 A TP 84 .5‐x‐1 3.15c 0.49 A 1010.91 1026.31 TP 85 2‐x‐2 3.85 6.20 A 1080.30 1071.39 TP 86 2‐x‐2 3.44 6.12 A 1122.97 1106.21 TP 87 2‐x‐2 2.97 6.56 B 993.47 1252.10 TP 88 2‐x‐2 2.93 4.80 B 953.20 1245.13 TP 89 2‐x‐2 2.88 5.24 B 947.57 1252.75 TP 90 1‐x‐1 2.14c 0.50 A 1004.67 1044.47 c TP 91 1‐x‐2 1.97 0.84 A 1009.09 1041.31 TP 92 1‐x‐2 2.67c 2.80 B 933.21 1238.43 TP 93 2‐x‐2 3.00 6.00 C 989.46 1160.03 TP 94 2‐x‐2 3.00 6.32 C 993.64 1166.72 c TP 95 1‐x‐2 2.11 2.80 C 1061.50 1208.35 TP 96 1‐x‐2 2.00c 2.68 C 1049.88 1220.03 TP 97 1‐x‐2 2.09c 2.80 C 1063.23 1207.27 Total 76.21 a Coordinates based on 1000N, 1000E steel rebar located at UTM E549051 N3371632 (Zone 16 WGS84). b Shovel test coordinates and elevations are center points; test pit coordinates and elevations are the southwest corner. c Ground surface after stripping.
91
Bayou Park, 8OK898
terminated without impacting the midden further. Excavated sediments were sifted through .64‐cm (1/4‐in) mesh screens. Most test units were excavated in Operation A (n=21). Six 1‐x‐2‐m units (TPs 58‐63) were excavated within the confines of demolished Buildings 588 and 591 prior to stripping (Figure 3.10). These served to better delineate the midden deposits in these two areas. A block of 20 1‐x‐1‐m units (TPs 64‐83) was laid out over a concentration of large pit features in Strip Area Ja in order to identify these in more detail, but only 10 of these were actually excavated (Figure 3.11). Test Pit (TP) 84 was a small (.5‐x‐1 m) unit excavated in the south wall of Strip Area K to investigate a sand midden stratum. TPs 85 and 86 were placed near PTA’s TPs 40 and 39, respectively, both of which contained shell, fired‐clay fragments, and lithics. TP85 was located between Buildings 626 and 632 and TP86 was excavated northwest of Building 633. In Operation B, a 1‐x‐2‐m unit (TP92) was excavated immediately to the south of PTA’s TP20 where a large shell‐filled pit and an abundance of lithic artifacts were identified in 2002. PTA’s TP20 was relocated through mechanical stripping; however, numerous utility lines in and around the location of TP20 limited where additional hand excavation could be conducted and only one unit was excavated. Additional test pits in Operation B were excavated in locations suggested by Eglin Archaeologist Joe Meyer. One 2‐x‐2‐m unit (TP87) was placed southeast of Building 635 and two more (TPs 88 and 89) were excavated southeast of Building 636. TPs 93 and 94 (2‐x‐2‐m) were excavated in Operation C just to the southeast of Building 634. This is near shovel tests and a 1‐x‐1‐m test unit (TP42) excavated previously by PTA that recovered shell and lithics. The last three units (TPs 95, 96, and 97) all measuring 1‐x‐2‐m, were excavated to the north of Building 635 in an effort to relocate and expose a black sandstone slab that was identified during monitoring of a water line in 2003. Using measurements from building corners provided by Eglin Archaeologist Joe Meyer, as well as GPR data collected from GPR Grid 44, two possible locations were identified. The hydraulic excavator was used to remove overburden in the area of the GPR anomaly and TP96 was excavated with negative results. The anomaly turned out to be large tree roots. Mechanical stripping also was performed in the measured location and fragments of black sandstone were encountered within a large modern pit. TPs 95 and 97 were excavated to better investigate the pit and to determine if any portion of the sandstone feature remained intact. All hand excavation was conducted using 10‐cm levels within observable strata and archaeological sediments were sifted through .64‐cm (1/4‐in) mesh screens. Modern fill was treated as a single stratigraphic unit and was excavated without screening. Vertical control was maintained with line levels using the unit’s highest corner as a datum plane. Elevations for each corner of each unit were established with the total station. All artifacts and bone recovered from the .64‐cm mesh screen were collected and placed in 4‐mil plastic zip‐lock bags. Complete quahog shells (Mercenaria spp.) also were collected for use in the seasonality analysis. Shell and charcoal samples for radiocarbon dating were piece‐plotted in situ, as were diagnostic artifacts when possible. 92
Research Design and Methods
Figure 3.10. Excavating TPs 59 and 61 inside demolished Building 588, Operation A.
Figure 3.11. Block excavation in Strip Area Ja.
93
Bayou Park, 8OK898
Figure 3.12. Documenting a shell‐filled pit feature (Feature 36) in Strip Area B.
Documentation included the use of standardized excavation forms for notes. Two wall profiles were photographed and drawn in each test unit. In addition, three 50‐x‐50‐cm column samples were excavated to recover fauna and botanical samples from controlled contexts within the midden. The column samples were taken from TP60, TP63, and the west wall of Strip Area L. These were collected as bulk samples from 10‐cm levels within observable strata. All features exposed through mechanical stripping or hand excavation were documented with photographs and plan view drawings, as well as x, y coordinates and elevation using the total station. Most features were bisected to document their profiles (Figure 3.12). Depending on the type of feature, bisection included hand excavation of one half of the feature and the sifting of excavated sediments through .32‐cm (1/8‐in) mesh screen; bisection with the hydraulic excavator; bisection with a shovel. A few exceptionally large features were sampled using small (25‐x‐25 cm) sampling units sifted through .32‐cm mesh. Profiles were drawn and photographed. Bulk samples were obtained from a representative sample of features for detailed faunal and botanical analysis. These bulk samples typically included the unexcavated half of hand‐excavated features. In addition, smaller soil samples were obtained from all pit features and many of the possible postmolds/postholes for possible chemical analysis. The sampling method included filling a 4‐x‐ 4‐in plastic zip‐lock bag (approximately 250 ml) with sediment from the center of the feature or from different lenses within a feature. A control sample also was collected from the
94
Research Design and Methods
surrounding matrix for possible sediment analysis. In situations where multiple features were present in close proximity to one another, a single control was collected equidistant from each feature in the group. The locations and elevations of soil and control samples were recorded with the total station and written in the field notes and on various feature maps.
LABORATORY METHODS Initial cleaning and processing of artifacts and other materials was conducted in the field using an on‐site building (Building 584) as temporary lab space. The indoor space was equipped with a computer station with printing capabilities, a standing rack of lipped .16‐cm (1/16‐in) aluminum drying screens, and space for rough sorting and storage. A washing and wet screening station was established outdoors with water provided by a spigot alongside the building and saw horses to support the aluminum drying screens (Figure 3.13). As materials were delivered to the field lab, they were sorted by provenience, then washed and allowed to air dry. Once dry, materials were separated by general artifact classes (lithics, ceramics [including baked‐clay objects], bone, shell, historic, and miscellaneous) and rebagged. All bags were assigned sequential Field Specimen (FS) numbers according to collection unit (i.e., by test unit, strip area, feature, etc.). Artifacts were counted and weighed using an Ohaus Scout Pro SP402, and all data and provenience information were entered into a Microsoft Access database developed by Eglin AFB cultural resource personnel. Preliminary processing of column samples and features also was conducted in the field. Column sample material was wet‐screened through .16‐cm (1/16‐inch) mesh and allowed to air dry before rebagging. For botanical analysis, a five‐liter sample was taken from the bulk samples of 26 selected features. The remainder of each feature was wet‐screened through .16‐cm mesh, air‐dried, and rebagged for potential faunal analysis. Reanalysis of previously recovered lithic artifacts and baked‐clay objects was performed at the Eglin curation facility prior to field work (March 2011). In‐house analyses of lithic, ceramic, bone, and shell artifacts and faunal remains recovered from form 2011‐2012 data recovery began after field work was completed in February 2012. Diagnostic artifacts that required detailed analysis or were planned for photographs were pulled from general provenience bags and placed in individual zip‐lock bags with the appropriate provenience information written on their exteriors. Chronometric samples, sediment samples, botanical samples, and clam shell samples were submitted to appropriate specialists. The reader is referred to individual chapters related to each of these materials for details of specific analysis methods.
NAGPRA‐RELATED HUMAN REMAINS Two human teeth were identified during analysis of the contents of Feature 44, a large refuse pit. Once the teeth were positively identified as human, they were separated from the rest of the feature material and placed in a zip‐lock plastic bag and then within a small acid‐free 95
Bayou Park, 8OK898
Figure 3.13. Water screening bulk feature sample through .16‐cm (1/16‐inch) mesh screen.
archival quality box, to await further notification from Eglin AFB regarding their treatment. The teeth were handled by only a single individual, SEARCH faunal analyst Brian Worthington. The Eglin Cultural Resources Manager, Mark Stanley, was notified by SEARCH Principal Investigator, Robert Austin, about the discovery and a written report was submitted. Consultation between Eglin AFB and NAGPRA representatives from the appropriate federally recognized tribes was initiated by Mr. Stanley. The teeth were wrapped in unbleached paper, packed in a tightly sealed container, and shipped to Mr. Stanley. After consultation with the tribes, the teeth were reburied on site. All unanalyzed features and column samples were scanned for any additional human remains. None were identified.
CURATION All cultural material, samples, and associated paperwork were prepared for long‐term storage in accordance with 36CFR79, Curation of Federally Owned and Administered Archaeological Collections and Eglin AFB’s Curation of Archaeological Collections Requirements (dated September 2011). All materials are permanently curated at Eglin Air Force Base under Task Order number CR‐11‐0051.
96
Research Design and Methods Appendix 3.1. Grid coordinates and elevation data for SEARCH hard points. Elevations Data Points N E Comments (m) Bldg. 591, SE corner 1020.28 1010.01 3.72 PTA point Bldg. 591, SW corner 1029.32 994.52 3.63 PTA point Bldg. 632, SE corner 1033.84 1084.66 4.14 PTA point Bldg. 632, SW corner 1048.19 1063.32 4.22 PTA point Bldg. 633, NE corner 1081.28 1171.02 3.92 PTA point Bldg. 633, NW corner 1111.86 1125.97 3.85 PTA point Bldg. 634, SE corner 987.98 1147.01 3.39 PTA point Bldg. 634, SW corner 1005.19 1121.44 3.36 PTA point Bldg. 635, outside side corner, NE 1017.87 1227.63 3.42 Bldg. 635, SE corner 986.81 1227.25 3.43 PTA point Bldg. 635, corner near 1000N/1200E 996.29 1213.22 3.45 Concrete Sewer Marker 996.00 964.00 3.30 PTA datum, NE corner of Inverness & Flagler; elevation is PTA's Fire hydrant 1, top 981.24 1270.56 3.80 Fire hydrant 2, top 925.71 1233.88 3.65 Fire hydrant 3, top, NE of Bldg. 633 1078.32 1173.68 4.35 Fire hydrant 4, top 1126.76 1126.02 4.51 Fire hydrant 5, top 1172.57 1104.33 4.01 Fire hydrant 6, top 1085.60 1021.90 4.26 Green electric box, NW corner 960.66 1233.08 3.22 Light Pole (LP) 1 914.13 1224.13 2.91 LP2 998.24 998.08 3.95 LP3‐Area A 1012.01 1023.91 4.05 LP3‐Area B 965.33 1271.80 3.10 LP5 1038.76 1064.23 4.09 LP6 994.36 1049.71 3.56 LP6 1047.65 1047.53 4.03 LP7 962.88 1042.32 3.39 LP8 1006.36 1087.43 3.46 LP9 1032.22 1069.90 3.85 LP13 970.77 1093.44 3.31 LP14 1012.79 1111.12 3.61 LP15 1034.15 1127.99 3.55 LP16 957.10 1123.22 3.44 LP18 962.90 1156.62 3.17 LP19 1002.53 1172.18 3.36 LP20 938.47 1165.55 3.08 LP21 971.59 1188.22 3.39 LP22 997.41 1200.96 3.40 LP23 928.23 1190.00 3.01 LP24 1070.83 1176.47 3.43
97
Bayou Park, 8OK898
Data Points
N
E
LP26 LP60(?), NE corner LP87(hand written) LP‐No # Manhole cover, SE corner concrete pad Power pole, spike at base
1022.55 939.28 1047.95 1133.22 912.28
1226.16 1252.28 1198.49 1090.28 1206.35
1068.33
1213.60
Elevations (m) 3.25 3.08 3.25 3.37 2.69
Comments
3.10 3.53 3.23 2.69 3.88 3.28 2.78 3.05 3.06 2.80 4.48 2.87 3.33
Top of Stake (TOS) SEARCH datum 1000.00 1000.00 PTA 1000N/1200E 1000.20 1199.74 TOS PTA 1000N/1300E 1000.45 1299.95 TOS Manhole Cover 1126.40 1122.31 N of Bldg. 633, PTA Datum 1 Manhole Cover 1069.73 1173.67 PTA Datum 2 Rebar, possible PTA datum 920.76 1248.94 Ground surface Telephone pole anchor, NE side 1147.16 1046.85 Temp BM‐PVC cap 978.12 1232.38 Temp BM‐PVC cap 1021.71 1255.35 Across from Bldg. 636 Water lift, top 1074.92 1035.63 Water meter, bolt on side 1035.26 1237.30 Water meter, bolt on side, N of 1099.11 1172.45 Bldg. 633 Water meter, top 1145.43 1127.28 4.80 White concrete wall, NW corner 1114.38 1018.63 3.51 NW side of Flagler St. a SEARCH coordinates based on 1000N, 1000E steel rebar located at UTM E549051 N3371632 (Zone 16 WGS84).
98
Type Number
Year
ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST ST
1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Appendix 3.2. Original and converted coordinates for PTA shovel tests and test pits. PTA Coordinates SEARCH Coordinatesa, b Notes North East North East ‐‐ ‐‐ 923.48 1344.64 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 904.71 1346.09 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 893.48 1324.81 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 907.75 1322.30 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 920.84 1323.23 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 948.07 1320.85 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 977.81 1295.74 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 951.11 1296.40 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 927.71 1299.57 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 906.17 1300.63 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 888.19 1300.76 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 886.61 1276.44 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 913.30 1275.64 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 932.07 1274.85 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 960.49 1274.06 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 984.94 1280.53 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 1015.11 1260.97 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 1007.31 1247.81 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 975.19 1249.13 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 955.23 1249.13 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 937.38 1249.79 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 915.84 1249.00 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 887.69 1248.73 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 863.90 1251.91 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 873.41 1228.77 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 915.18 1221.90 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 936.59 1221.37 Location estimated from Campbell and Meyer 1993:Figure 3 ‐‐ ‐‐ 953.38 1221.24 Location estimated from Campbell and Meyer 1993:Figure 3
Type Number ST 29 ST 30 ST 31 ST 32 ST 33 ST 34 ST 35 ST 36 ST 43 ST 44 ST 45 ST 46 ST 47 ST 48 ST 49 ST 50 ST 51 ST 52 ST 53 ST 54 ST 55 ST 56 ST 57 ST 58 ST 59 ST 60 ST 61 ST 62 ST 63 ST 64
Year 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993 1993
PTA Coordinates SEARCH Coordinatesa, b ‐‐ ‐‐ 935.01 1207.89 ‐‐ ‐‐ 915.58 1207.23 ‐‐ ‐‐ 891.91 1206.04 ‐‐ ‐‐ 847.44 1184.34 ‐‐ ‐‐ 842.76 1155.44 ‐‐ ‐‐ 825.24 1089.80 ‐‐ ‐‐ 854.50 1002.10 ‐‐ ‐‐ 874.60 936.10 ‐‐ ‐‐ 997.40 1203.53 ‐‐ ‐‐ 1004.00 1172.60 ‐‐ ‐‐ 982.06 1172.34 ‐‐ ‐‐ 962.77 1174.32 ‐‐ ‐‐ 941.75 1152.64 ‐‐ ‐‐ 962.37 1152.77 ‐‐ ‐‐ 937.65 1130.96 ‐‐ ‐‐ 957.74 1132.29 ‐‐ ‐‐ 975.45 1132.95 ‐‐ ‐‐ 974.00 1096.20 ‐‐ ‐‐ 997.92 1096.86 ‐‐ ‐‐ 1025.31 1102.68 ‐‐ ‐‐ 1056.37 1105.45 ‐‐ ‐‐ 1053.08 1132.95 ‐‐ ‐‐ 982.88 1080.74 ‐‐ ‐‐ 1003.90 1081.27 ‐‐ ‐‐ 995.97 1056.81 ‐‐ ‐‐ 1038.53 1056.15 ‐‐ ‐‐ 1008.52 1033.02 ‐‐ ‐‐ 1071.04 1058.40 ‐‐ ‐‐ 1127.35 1055.09 ‐‐ ‐‐ 1128.94 1105.45
Notes Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3
Type Number ST 65 ST 66 ST 67 ST 68 ST 69 ST 77 ST 78 ST 79 ST 80 ST 81 ST 82 ST 83 ST 84 ST 85 ST 86 ST 87 ST 88 ST 89 ST 90 ST 91 ST 103 ST 104 ST 105 ST 96‐1 ST 96‐2 ST 96‐3 ST 96‐4 ST 96‐5 ST 96‐6 ST 96‐7
Year 1993 1993 1993 1993 1993 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1994 1996 1996 1996 1996 1996 1996 1996
PTA Coordinates SEARCH Coordinatesa, b ‐‐ ‐‐ 1019.10 992.05 ‐‐ ‐‐ 892.17 1072.26 ‐‐ ‐‐ 904.20 1051.62 ‐‐ ‐‐ 928.28 1018.02 ‐‐ ‐‐ 970.48 957.17 ‐‐ ‐‐ 1044.48 1490.24 ‐‐ ‐‐ 1025.51 1447.24 ‐‐ ‐‐ 1000.90 1422.63 ‐‐ ‐‐ 967.69 1382.61 ‐‐ ‐‐ 934.20 1357.14 ‐‐ ‐‐ 968.87 1327.46 ‐‐ ‐‐ 996.04 1298.42 ‐‐ ‐‐ 1034.02 1265.00 ‐‐ ‐‐ 1071.76 1231.61 ‐‐ ‐‐ 1107.05 1204.23 ‐‐ ‐‐ 1093.28 1335.82 ‐‐ ‐‐ 1103.53 1359.97 ‐‐ ‐‐ 1162.50 1329.10 ‐‐ ‐‐ 1182.30 1360.40 ‐‐ ‐‐ 1194.60 1383.10 ‐‐ ‐‐ 1102.50 1463.20 ‐‐ ‐‐ 1066.49 1376.50 ‐‐ ‐‐ 1104.79 1198.11 ‐‐ ‐‐ 1030.40 1244.10 ‐‐ ‐‐ 1039.10 1227.90 ‐‐ ‐‐ 1059.70 1217.20 ‐‐ ‐‐ 1061.70 1199.20 ‐‐ ‐‐ 1107.80 1188.40 ‐‐ ‐‐ 1131.00 1166.60 ‐‐ ‐‐ 1131.70 1156.60
Notes Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Campbell and Meyer 1993:Figure 3 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Location estimated from Mathews et al. 1994:Figure 35 Estimated location based on Mikell et al. 1996:Figure 14 Estimated location based on Mikell et al. 1996:Figure 14 Estimated location based on Mikell et al. 1996:Figure 14 Estimated location based on Mikell et al. 1996:Figure 14 Estimated location based on Mikell et al. 1996:Figure 14 Estimated location based on Mikell et al. 1996:Figure 14 Estimated location based on Mikell et al. 1996:Figure 14
Type Number ST 96‐8 ST 96‐9 ST 109 ST 110 ST 111 ST 112 ST 113 ST 114 ST 115 ST 116 ST 117 ST 118 ST 119 ST 120 ST 121 ST 122 ST 123 ST 124 ST 125 ST 126 ST 127 ST 128 ST 129 ST 130 ST 131 ST 132 ST 133 ST 134 ST 135 ST 136
Year 1996 1996 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 2002 2002 2002 2002 2002 2002 2002
PTA Coordinates SEARCH Coordinatesa, b ‐‐ ‐‐ 888.90 1225.20 ‐‐ ‐‐ 889.70 1192.10 ‐‐ ‐‐ 1003.13 974.80 ‐‐ ‐‐ 1007.95 977.73 ‐‐ ‐‐ 1013.82 981.38 ‐‐ ‐‐ 1020.17 984.90 ‐‐ ‐‐ 1025.93 988.78 ‐‐ ‐‐ 1021.17 995.94 ‐‐ ‐‐ 1015.77 992.76 ‐‐ ‐‐ 1009.42 988.89 ‐‐ ‐‐ 1003.90 985.01 ‐‐ ‐‐ 998.85 981.60 ‐‐ ‐‐ 994.54 987.46 ‐‐ ‐‐ 999.83 991.11 ‐‐ ‐‐ 1005.23 994.87 ‐‐ ‐‐ 1011.81 999.10 ‐‐ ‐‐ 1017.22 1002.74 ‐‐ ‐‐ 1013.45 1009.10 ‐‐ ‐‐ 1002.00 999.80 ‐‐ ‐‐ 998.55 998.41 ‐‐ ‐‐ 989.38 995.01 ‐‐ ‐‐ 1008.18 1000.65 ‐‐ ‐‐ 995.61 1003.23 1000.00 1000.00 1004.95 999.39 1005.00 1013.00 1008.47 1018.11 1005.00 1022.00 1008.05 1023.30 1023.00 1040.00 1023.72 1038.23 1040.00 1052.00 1040.72 1050.23 1056.00 1066.00 1054.92 1060.95 1031.00 1084.00 1076.32 1074.71
Notes Estimated location based on Mikell et al. 1996:Figure 14 Estimated location based on Mikell et al. 1996:Figure 14 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 Estimated from PTA field notes, 11/19/1999 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002
Type Number ST 137 ST 138 ST 139 ST 140 ST 141 ST 142 ST 143 ST 144 ST 145 ST 146 ST 147 ST 148 ST 149 ST 150 ST 151 ST 152 ST 153 ST 154 ST 155 ST 156 ST 157 ST 158 ST 159 ST 160 ST 161 ST 162 ST 163 ST 164 ST 165 ST 166
Year 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002
PTA Coordinates SEARCH Coordinatesa, b 1120.00 1112.00 1114.05 1101.02 1047.00 1143.00 1148.04 1129.23 1131.00 1156.00 1132.04 1142.23 1114.00 1171.00 1115.04 1157.23 1098.00 1186.00 1099.04 1172.23 1084.00 1203.00 1085.04 1189.23 1071.00 1219.00 1072.04 1205.23 1056.00 1234.00 1057.04 1220.23 1040.00 1248.00 1042.37 1232.13 1023.00 1265.00 1024.04 1251.23 1010.00 1280.00 1011.04 1266.23 974.00 1288.00 989.07 1267.61 956.00 1276.00 965.98 1261.16 939.00 1262.00 948.98 1247.16 925.00 1949.00 940.07 1228.61 922.00 1240.00 928.82 1230.05 932.00 1228.00 938.82 1218.05 933.00 1256.00 942.98 1241.16 949.00 1269.00 958.98 1254.16 966.00 1282.00 975.98 1267.16 995.00 1290.00 1010.07 1269.61 927.00 1228.00 933.82 1218.05 927.00 1232.00 933.82 1222.05 932.00 1224.00 938.82 1214.05 1006.00 1008.00 1007.99 1009.79 1009.00 1030.00 1009.72 1028.23 1014.00 1033.00 1014.72 1031.23 1019.00 1036.00 1019.19 1034.86 1027.00 1043.00 1027.72 1041.23 1032.00 1046.00 1032.72 1044.23
Notes PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002
Type Number ST 167 ST 168 ST 169 ST 170 ST 171 ST 172 ST 173 ST 174 ST 175 ST 176 ST 177 ST 178 ST 179 ST 180 ST 181 ST 182 ST 183 ST 184 ST 185 ST 186 ST 187 ST 188 ST 189 ST 190 ST 191 ST 192 ST 193 ST 194 ST 195 ST 196
Year 2002 2002 2002 2002 2002 2002 2002 2002 2007 2007 2007 2007 2007 2007 2007 2007 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008
PTA Coordinates SEARCH Coordinatesa, b 1009.00 1035.00 1013.31 1032.92 1009.00 1025.00 1013.20 1028.10 1027.00 1048.00 1024.33 1042.95 1027.00 1038.00 1028.04 1037.71 933.00 1256.00 942.98 1241.16 943.00 1265.00 952.98 1250.16 952.00 1272.00 961.98 1257.16 929.00 1253.00 936.98 1238.16 ‐‐ ‐‐ 965.67 1078.15 ‐‐ ‐‐ 954.73 1085.98 ‐‐ ‐‐ 987.62 954.36 ‐‐ ‐‐ 927.79 1046.34 ‐‐ ‐‐ 895.05 1094.76 ‐‐ ‐‐ 931.82 1164.14 ‐‐ ‐‐ 993.47 1212.83 ‐‐ ‐‐ 990.56 1217.06 ‐‐ ‐‐ 1024.62 1267.43 ‐‐ ‐‐ 1050.55 1244.41 ‐‐ ‐‐ 1078.86 1221.39 ‐‐ ‐‐ 1107.05 1204.23 ‐‐ ‐‐ 1130.60 1176.90 ‐‐ ‐‐ 1149.51 1183.29 ‐‐ ‐‐ 1156.91 1206.58 ‐‐ ‐‐ 1129.40 1229.59 ‐‐ ‐‐ 1103.73 1252.88 ‐‐ ‐‐ 1076.48 1275.37 ‐‐ ‐‐ 1049.23 1298.12 ‐‐ ‐‐ 1028.86 1290.45 ‐‐ ‐‐ 1052.14 1270.08 ‐‐ ‐‐ 1077.54 1248.12
Notes PTA field notes 8/28/2002; relocated SEARCH 2012 PTA field notes 8/28/2002; relocated SEARCH 2012 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 PTA field notes 8/28/2002 Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio
Type Number ST 197 ST 198 ST 199 ST 200 ST 201 ST 202 ST 203 ST 204 ST 205 ST 206 ST 207 ST 208 ST 209 ST 210 ST 211 ST 212 ST 213 ST 214 ST 215 ST 216 ST 217 ST 218 ST 219 ST 220 ST 221 ST 222 ST 223 ST 224 ST 225 ST 226
Year 2008 2008 2008 2008 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009
PTA Coordinates SEARCH Coordinatesa, b ‐‐ ‐‐ 1103.73 1224.83 ‐‐ ‐‐ 1130.46 1202.08 982.00 1232.00 982.80 1232.90 986.00 1226.00 986.80 1226.90 1173.50 1116.50 1174.30 1117.44 1144.50 1133.50 1145.30 1134.44 1113.50 1166.50 1114.30 1167.44 1096.50 1150.00 1097.30 1150.94 1093.00 1149.50 1093.80 1150.44 1094.50 1152.50 1095.30 1153.44 1090.50 1151.50 1091.30 1152.44 1011.50 1169.00 1012.30 1169.94 1011.50 1163.50 1012.30 1164.44 1012.00 1159.50 1012.80 1160.44 1001.50 1137.50 1002.30 1138.44 1004.00 1137.30 1004.80 1138.24 1006.90 1141.00 1007.70 1141.94 1011.20 1141.70 1012.00 1142.64 1044.90 1163.80 1045.70 1164.74 1042.50 1166.50 1043.30 1167.44 1039.00 1160.50 1039.80 1161.44 1036.10 1160.30 1038.90 1161.24 1031.90 1157.40 1032.70 1158.34 1031.60 1156.30 1032.40 1156.24 1076.70 1157.90 1077.50 1158.84 1074.20 1154.50 1075.00 1155.44 1095.60 1121.80 1096.40 1122.74 1094.80 1127.10 1095.60 1128.04 1092.20 1130.90 1093.00 1131.84 1080.70 1100.30 1081.50 1101.24
Notes Estimated from Morehead, Campbell, and Aubuchon 2008: folio Estimated from Morehead, Campbell, and Aubuchon 2008: folio PTA field notes 2008 PTA field notes 2008 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152
Type Number ST 227 ST 228 ST 229 ST 230 ST 231 ST 232 ST 233 ST 234 ST 235 ST 236 ST 237 ST 238 ST 239 ST 240 ST 241 ST 242 ST 243 ST 244 ST 245 ST 246 ST 247 ST 248 ST 249 ST 250 ST 251 ST 252 ST 253 ST 254 ST 255 ST 256
Year 2009 2009 2009 2009 2009 2009 2009 2009 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010
PTA Coordinates SEARCH Coordinatesa, b 1076.60 1097.60 1077.40 1098.54 1072.80 1095.00 1073.60 1095.94 1055.60 1081.70 1056.40 1082.64 1052.80 1081.80 1053.60 1082.74 1072.80 1096.10 1073.60 1079.64 1071.50 1080.70 1072.30 1082.04 1076.20 1084.30 1077.00 1085.24 1077.80 1081.30 1078.60 1082.24 988.30 1239.20 985.01 1244.29 984.40 1242.20 980.14 1248.55 980.20 1244.90 974.38 1251.92 976.00 1242.00 968.81 1251.35 972.10 1239.40 964.91 1248.68 967.90 1236.90 960.95 1245.71 964.20 1234.70 957.55 1243.07 963.30 1231.10 958.09 1238.11 965.30 1228.20 962.36 1231.98 969.10 1225.90 971.12 1228.82 966.50 1224.10 966.74 1226.26 963.50 1222.20 962.11 1223.46 975.60 1246.40 968.47 1255.25 972.10 1244.10 964.51 1253.49 966.70 1243.80 960.47 1253.79 964.30 1238.90 958.09 1250.26 954.80 1235.80 951.94 1235.31 959.30 1229.40 952.20 1230.10 942.10 1229.90 940.95 1227.47 932.70 1232.70 929.36 1232.85 938.30 1238.50 938.37 1240.04 944.80 1244.80 945.56 1245.52
Notes Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C
Type Number TP 1 TP 2 TP 3 TP 4 TP 5 TP 6 TP 7 TP 8 TP 9 TP 10 TP 11 TP 12 TP 13 TP 14 TP 15 TP 16 TP 17 TP 18 TP 19 TP 20 TP 21 TP 22 TP 23 TP 24 TP 25 TP 26 TP 27 TP 28 TP 29 TP 30
Year 1996 1996 1996 1997 1997 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 2002 2002 2007 2007 2007 2007 2007 2007 2007 2007 2007 2007
PTA Coordinates SEARCH Coordinatesa, b ‐‐ ‐‐ 1004.46 1242.95 ‐‐ ‐‐ 947.82 1272.90 ‐‐ ‐‐ 949.33 1259.51 ‐‐ ‐‐ 883.55 1196.06 ‐‐ ‐‐ 861.21 1180.95 ‐‐ ‐‐ 1002.00 999.80 ‐‐ ‐‐ 1001.40 999.10 ‐‐ ‐‐ 999.10 1002.70 ‐‐ ‐‐ 1000.50 1002.70 ‐‐ ‐‐ 1001.40 1000.50 ‐‐ ‐‐ 1000.60 1000.00 ‐‐ ‐‐ 1002.00 998.40 ‐‐ ‐‐ 1003.50 1000.00 ‐‐ ‐‐ 1000.00 999.30 ‐‐ ‐‐ 1000.00 1000.70 ‐‐ ‐‐ 1004.20 1000.70 ‐‐ ‐‐ 1004.90 1000.00 ‐‐ ‐‐ 997.70 1002.70 1026.00 1041.00 1024.55 1040.55 927.00 1252.70 933.63 1238.30 ‐‐ ‐‐ 975.46 1079.04 ‐‐ ‐‐ 945.64 1061.01 ‐‐ ‐‐ 941.59 1060.98 ‐‐ ‐‐ 1081.94 1065.15 ‐‐ ‐‐ 1095.92 1096.57 ‐‐ ‐‐ 1081.47 1124.22 ‐‐ ‐‐ 1054.36 1154.04 ‐‐ ‐‐ 993.03 1209.19 ‐‐ ‐‐ 1080.90 1128.70 ‐‐ ‐‐ 964.70 1067.30
Notes Estimated location based on PTA field notes 1996 Estimated location based on PTA field notes 1996 Estimated location based on PTA field notes 1996 Estimated location based on Campbell and Mathews 1997 Estimated location based on Campbell and Mathews 1997 Relocated SEARCH 2012 Relocated SEARCH 2012 Relocated SEARCH 2012 Relocated SEARCH 2012 Relocated SEARCH 2012 Relocated SEARCH 2012 Calculated from SEARCH relocation data Relocated SEARCH 2012 Calculated from SEARCH relocation data Calculated from SEARCH relocation data Relocated SEARCH 2012 Relocated SEARCH 2012 Calculated from SEARCH relocation data PTA field notes 8/28/2002; relocated SEARCH 2012 PTA field notes 8/28/2002; relocated SEARCH 2012 Estimated from PTA field notes 2007 Estimated from PTA field notes 2007 Estimated from PTA field notes 2007 Estimated from PTA field notes 2007 Estimated from PTA field notes 2007 Estimated from PTA field notes 2007 Estimated from PTA field notes 2007 Estimated from PTA field notes 2007 Estimated location based on Morehead et al. 2008:Appendix A Estimated location based on Morehead et al. 2008:Appendix A
Type Number Year PTA Coordinates SEARCH Coordinatesa, b Notes TP 31 2008 1024.00 1223.00 1029.64 1225.74 PTA field notes 2008 TP 32 2008 1024.00 1219.00 1030.79 1219.90 PTA field notes 2008 TP 33 2008 1027.00 1225.00 1034.36 1229.01 PTA field notes 2008 TP 34 2008 1028.00 1221.00 1037.32 1221.44 PTA field notes 2008 TP 35 2008 1022.00 1221.00 1026.14 1223.05 PTA field notes 2008 TP 36 2008 985.00 1229.00 990.13 1234.69 PTA field notes 2008 TP 37 2008 987.00 1225.00 987.80 1225.90 PTA field notes 2008 TP 38 2009 ‐‐ ‐‐ 1052.14 1270.08 Estimated location based on Bourgeois et al. 2009:Figure 22 TP 39 2009 1126.50 1116.50 1127.30 1117.44 Campbell, Morehead, et al. 2009:151‐153 TP 40 2009 1081.00 1076.00 1081.80 1076.94 Campbell, Morehead, et al. 2009:151‐153 TP 41 2009 1098.50 1151.50 1099.30 1152.44 Campbell, Morehead, et al. 2009:151‐153 TP 42 2009 1010.00 1172.00 1010.80 1172.94 Campbell, Morehead, et al. 2009:151‐153 TP 43 2009 1005.30 1139.10 1006.10 1140.04 Campbell, Morehead, et al. 2009:151‐153 TP 44 2009 1034.30 1157.70 1035.10 1158.64 Campbell, Morehead, et al. 2009:151‐153 TP 45 2009 1079.20 1149.70 1080.00 1150.64 Campbell, Morehead, et al. 2009:151‐153 TP 46 2009 1093.10 1127.40 1093.90 1128.34 Campbell, Morehead, et al. 2009:151‐153 TP 47 2009 ‐‐ ‐‐ 1075.00 1097.04 Campbell, Morehead, et al. 2009:151‐153 TP 48 2009 ‐‐ ‐‐ 1053.90 1081.64 Campbell, Morehead, et al. 2009:151‐153 TP 49 2009 1112.50 1166.50 1113.30 1167.44 Campbell, Morehead, et al. 2009:151‐153 TP 50 2010 940.4 1220.60 939.25 1219.53 Morehead et al. 2011:Appendix C TP 51 2010 939.7 1215.20 938.00 1215.79 Morehead et al. 2011:Appendix C TP 52 2010 960.9 1233.00 954.32 1241.73 Morehead et al. 2011:Appendix C TP 53 2010 966 1229.50 963.45 1236.13 Morehead et al. 2011:Appendix C TP 54 2010 967.5 1231.50 966.25 1239.30 Morehead et al. 2011:Appendix C TP 55 2010 992.2 1234.50 991.17 1242.00 Morehead et al. 2011:Appendix C TP 56 2010 967.5 1237.00 966.56 1237.48 Morehead et al. 2011:Appendix C TP 57 2010 968.5 1231.00 967.47 1237.96 Morehead et al. 2011:Appendix C a Because it was not always clear which corner of a shovel test or test pit was used to designate a specific unit’s coordinates, all SEARCH coordinates are assumed to be center points. b SEARCH coordinates based on 1000N, 1000E steel rebar located at UTM E549051 N3371632 (Zone 16 WGS84).
4
STRATIFICATION AND CHRONOLOGY
In this chapter we discuss the stratification and radiometric chronology at Bayou Park. The basic sequence of strata at the site was established during PTA’s block excavation in 1998 and a subsequent survey of a utility line corridor in 2002 (Campbell et al. 1998; Meyer et al. 2002). Samples for radiocarbon assays were collected during 1998 and 2002 and these provided a basic time frame for the site’s primary occupation. SEARCH’s data recovery project in 2011‐2012 served to refine and build on these initial interpretations. Long profiles in several north‐south and east‐west trenches allowed for the development of a detailed site‐wide stratigraphy and a better understanding of site formation processes and the relationships between midden deposition and natural soil deposition and geomorphology. Twelve radiocarbon assays, coupled with the four dates obtained by PTA, indicate occupation between 4210‐3690 BP (3210‐1817 cal BC) with additional occupations at 1990 BP (cal 48 BC‐AD 72) and 250 BP (cal AD 1522‐modern). These dates correspond primarily with the Late Archaic period, or what is referred to by Thomas and Campbell (1991, 1993) as the Gulf Formational period; one date corresponds with the Middle Woodland period, and one is possibly post‐Contact. Temporally diagnostic artifacts (baked‐ clay objects, microliths, Florida Archaic Stemmed projectile points) indicate that the earliest occupation was by participants in the Elliott’s Point culture. A few sherds of Weeden Island and Fort Walton pottery indicate later occupations during the Woodland and Mississippi periods, while deeply buried lithic artifacts suggest a pre‐Late Archaic occupation. None of these approached the intensity of the Elliott’s Point occupation, however.
DESCRIPTION OF STRATA Prior to entering the field, the original strata descriptions from various PTA reports were summarized and a master sequence was prepared for use during excavation. This master sequence, which is shown in Table 4.1, essentially followed PTA’s descriptions with the following modifications. The midden stratum was divided into two major substrata IVa (sand midden) and IVb (shell midden) to correspond with PTA’s Ab1 and Ab2 designations. During excavation, a lighter, less organic, but still identifiable stratum was identified and labeled IVc. Stratum V, the immediate submidden stratum was retained, but the various soil horizons visible below this stratum, particularly in Operations B and C, were collapsed into a single stratum designation, Stratum VI. The reason for this last decision is that horizon formation within the site’s sandy non‐midden sediments is typically the result of pedogenesis rather than anthropomorphic processes. Therefore, any artifacts found within these lower strata either were deposited much earlier than the formation of visible horizons, or had migrated to their current position via bioturbation (see Leigh 1998, 2001; Thulman 2012; Wilder et al. 2001). In either case, the separate excavation of these horizons
Bayou Park, 8OK898 Table 4.1. Strata correlations between representative shovel tests and test pits, midden area. SEARCH Descriptions
PTA Strata Descriptions TP9 Depth in Horizon cmbs 0‐12 Fill
ST163 Depth in Horizon cmbs 0‐22 Fill
TP19 Depth in Horizon cmbs 0‐50 Fill
Strata
Horizons
I
O/Ap/Fill
II
Ab
12‐21
Ap
22‐38
Ab
50‐58
Ab
III
AC/(Fill?)
21‐31
AC
‐‐
‐‐
58‐108
AC
IVa
Ab
31‐58
Ab1
‐‐
‐‐
108‐150+
Ab
IVb
Ab
32‐53
Ab2
38‐58
Ab2
‐‐
‐‐
IVc
Ab C1 (leach zone) C2
‐‐
‐‐
‐‐
‐‐
‐‐
‐‐
38‐63
Ab3
58‐63
Ab3
‐‐
‐‐
55‐70
C
63‐100
C
133‐150+
C
V VI
as stratigraphically distinct collection units would not have resulted in useful contextual information. Stratum VI was, however, excavated in arbitrary 10‐cm levels (see Chapter 3). The strata designations in the first column of Table 4.1 were used throughout the field work and are referenced throughout the remainder of the report. Descriptions of these stratigraphic units are presented below. Profile summaries are presented in Figures 4.1 and 4.2 and representative test pit profiles are presented in Figures 4.3‐4.9. Detailed profiles of strip areas are provided in Appendix 4.
Stratum I Stratum I is a composite of the modern organic (O) and mineral (Ap) horizons and the underlying fill, which for the purposes of excavation were lumped together as a single stratum. The modern topsoil (Ia) consisted of very dark gray (10YR3/1) to gray (10YR5/1), loose, medium‐grained sand mixed with organic material and small grass roots. The fill (Ib) is primarily a homogeneous clay or sandy clay that ranges in color from red (2.5YR4/6, 5/6, 5/8) to reddish‐brown (2.5YR4/8) to reddish‐yellow (5YR7/8). It is extremely hard and compact, especially when dry, has a blocky structure, and an abrupt lower boundary. The thickness of the clay fill is variable ranging from a few centimeters to as much as a meter; however, on average, it is between 20 and 40 cm thick. The clay fill was commonly encountered in Operations A and C, and occasionally in Operation B. Lenses of sand fill (Ic) and isolated fill deposits also occur which appear to be related to specific construction events or the installation of utilities. These fill deposits usually consist of sand, sand and rubble, sand and small rounded pebbles, sand and clay fragments, or shell. Color is extremely variable and includes very pale brown (10YR7/4, 8/2), pale brown (10YR6/3), brown (10YR4/3, 5/3), dark yellowish‐brown (10YR4/4, 4/6), yellowish‐brown (10YR5/6), light yellowish‐brown (10YR6/4), very dark gray (10YR4/2), very dark grayish‐ brown (10YR3/2), and white (10YR8/1). PTA reports modern intrusions containing sand fill
110
Figure 4.1. Generalized north‐south profile through Operation A based on profiles in Strip Areas E, D, and K. Vertical scale exaggerated and distances between profiled sections are not to scale.
Stratification and Chronology
111
Figure 4.2. Generalized east‐west profile through Operations A, C, and B based on profiles in Strip Areas. Vertical scale exaggerated and distances between profiled sections are not to scale.
Bayou Park, 8OK898
112
Stratification and Chronology
Figure 4.3. East wall profile, Test Pit 59, Operation A.
Figure 4.4. East wall profile, Test Pit 60, Operation A.
113
Bayou Park, 8OK898
Figure 4.5. West wall profile, Test Pit 62, Operation A.
Figure 4.6. West wall profile, Test Pit 63, Operation A.
114
Stratification and Chronology
Figure 4.7. South wall profile, Test Pit 85, Operation A.
extending to depths greater than one meter in isolated areas, particularly near buildings or within their footprints (e.g., Campbell, Morehead, et al. 2009:77‐90; Morehead, Campbell, and Aubuchon 2008:64‐74), but most of the isolated fill deposits encountered during the 2011‐2012 project were not as extensive. That exception was to the north of Building 635 where TPs 95, 96, and 97 encountered sand fill extending to about 95 cmbs. Fill deposits in Operation B tended to consist of sand, sand and gravel, or sand and shell (Rangia cuneata) rather than red clay (see Figure 4.8). As Stratum I was not screened when hand excavated, it is not possible to discuss artifact content except for materials that were observed as work progressed. These typically included pieces of building rubble, cement fragments, glass fragments and occasional whole bottles, PVC pipe fragments, bullets and bullet casings, iron fragments, nails, and occasional marine shells presumably mixed into the fill from the underlying midden.
115
Bayou Park, 8OK898
Figure 4.8. South wall profile, Test Pit 88, Operation B.
Stratum II Stratum II was not consistently observed, perhaps because it was removed in some areas during site grading and the deposit of fill. Where present in Operations A and C, it consisted of light gray (10YR6/1), grayish‐brown (10YR5/4), or black (10YR2/1) fine to medium sand or loamy sand immediately underlying the reddish clay. It was usually rather thin (2‐10 cm), but sometimes was as much as 20‐25 cm in thickness. It was typically loose to slightly compact with a relatively clear lower boundary. Stratum II contained occasional modern artifacts (glass, metal, rubble) and marine shell fragments, as well as evidence of burning (see Figure 4.7). In Operations B and C, it was more variable and occasionally displayed a distinct buried surface horizon (see Figure 4.9). More common, however, was a zone of sand lenses of various colors or a mottled sand stratum.
Stratum III This stratum occurred only in Operation A. It typically consisted of yellowish‐brown (10YR5/4, 5/6) to dark yellowish‐brown (10YR4/4) fine to medium sand, occasionally mottled, sometimes with very dark grayish‐brown (10YR3/2) or dark brown (10YR3/3) fine to medium sand. It varied in thickness from a few centimeters to as much as 45 cm, and in some places was entirely absent. Stratum III contained prehistoric artifacts and faunal material, but in relatively low quantities compared to Stratum IV. It also contained pieces of building rubble in a few locations.
116
Stratification and Chronology
Figure 4.9. South wall profile, Test Pit 94, Operation C.
Stratum IV Stratum IV is the prehistoric midden, an anthropogenically derived deposit that consists of organic‐stained sediments, marine shell, animal bone, and artifacts. The thickness of this stratum varied from between 20 and 45 cm. PTA divided Stratum IV into two substrata, one containing mostly sand and the other dominated by shell, and this practice was continued during the SEARCH project. Stratum IVa was characterized by very dark brown (10YR2/2), brown (10YR3/3, 10YR4/3), dark grayish‐brown (10YR4/2), and very dark gray (10YR3/1), slightly compact fine to medium sand while Stratum IVb was dark brown (10YR2/2), black (10YR2/1), or very dark gray (10YR3/1) in color and composed of slightly compact fine to medium sand and abundant marine shell. The lower boundaries of these substrata tended to be abrupt and irregular. The mechanical stripping revealed that Stratum IVb was not a true stratum but instead consisted of discontiguous deposits of shell within the larger
117
Bayou Park, 8OK898
midden deposit. This was seen in both plan view and profiles (see Figures 4.1, 4.2, 4.4, and 4.6). Stratum IVc was identified in TPs 58, 59 (see Figure 4.3) and 61 and in the floor of Strip Area C. This substratum was characterized by brown (10YR5/3), yellowish‐brown (10YR5/4), and gray (10YR5/1), slightly compact fine to medium sand with observable charcoal flecking. Again, this was not a distinct stratum per se, but appeared to be an extension of the midden deposit restricted to its periphery where the discard of organic materials (and hence organic staining) was not as intense. The lower boundary of Stratum IVc was irregular and conformable with the underlying Stratum VI. Thickness varied from 10 to 25 cm.
Stratum V Stratum V was present only in Operation A, usually in association with Stratum IV (see Figures 4.1, 4.2, 4.4, and 4.6). It is characterized by a mottled appearance with dark grayish‐brown (10YR4/2) or dark yellowish‐brown (10YR4/4), slightly compact fine to medium sand mottled with yellowish‐brown (10YR5/4) sand. Marine shell and artifacts are present in this stratum, but are not as abundant as in Stratum IV. Beneath Stratum IVc, it consists of gray (10YR5/1) or yellowish‐brown (10YR5/4) sand with charcoal flecking (see Figure 4.3). Stratum V is generally about 20 cm thick with a lower boundary that is relatively smooth and conformable with Stratum VI. It is best observed in areas where Stratum IV is well developed. This and the conformable nature of the lower boundary and yellowish‐brown mottling suggest that Stratum V represents a zone of organic and mineral leaching from the overlying midden into the underlying yellowish‐brown sediments of Stratum VI.
Stratum VI As mentioned above, Stratum VI represents a composite of the naturally occurring soils within the Bayou Park site. In Operation A, it underlies Stratum IV (the midden) and Stratum V while in Operation C it underlies Stratum I. It is characterized by yellowish‐brown (10YR5/4), light yellowish‐brown (10YR6/4), brownish‐yellow (10YR6/6), and yellow (10YR7/6) medium‐grained, slightly compact sand. Small pockets of white (10YR8/1) clay were sometimes encountered at depths of 1.25 to 1.5 meters below ground surface. This horizon sequence is similar to Foxworth soil as described by the National Resource Conservation Service (https://soilseries.sc.egov.usda.gov/OSD_Docs/F/FOXWORTH.html) (NRCS 2012). In Operation A, this stratum contained numerous features but artifact content was low except near its contact with Strata IV or V. The base of this stratum was not reached in Operation A, but thicknesses of the exposed sediments ranged between 60 and 120 cm. In Operation B, Stratum VI immediately underlies Stratum I and/or Stratum II and exhibits the horizonation associated with Kureb (https://soilseries.sc.egov.usda.gov/OSD Docs/K/KUREB.html) or Resota (https://soilseries.sc.egov.usda.gov/OSD_Docs/R/RESOTA. html) soils (NRCS 2012). This stratum consisted of a relatively thin (15‐30 cm) E horizon of
118
Stratification and Chronology
white (10YR8/1) to light gray (10YR7/1) fine to medium sand underlain by an poorly developed B horizon or B/C horizon (10YR4/6, dark yellowish‐brown sand) about 15‐40 cm thick, followed by a thick (50‐90 cm) C horizon (10YR6/6, brownish‐yellow sand). Artifacts contained within this stratum in Operation B consisted primarily of lithic waste flakes and occasional tools. Only a few features were identified but no midden (Stratum IV).
Laboratory Analysis of Sediment Samples In addition to the field descriptions, soil samples were obtained from Strata III through VI in TPs 58 and 60 in order to better characterize the sediments that make up the site. All samples were obtained from the east wall profile. The samples from TP 60 were subsequently examined in the laboratory under a 70x Stereo Zoom microscope. After describing their content and color while wet, the samples were weighed and placed in a conventional oven for 16 hours at a temperature of 230˚ F. After removal from the oven, the samples were reweighed and the difference in weight between wet and dry samples was recorded as a measure of the moisture content of the sediment. Each sample was tested with dilute (10%) HCL to test for the presence of carbonates. This information is presented in Table 4.2. For the most part, the laboratory descriptions and analysis are consistent with the field descriptions in terms of color and content. The main difference is in the characterization of the sand fraction. Field observations consistently classified the sand as fine to medium‐ grained, but examination under a microscope indicated a substantial amount of coarse and very coarse quartz grains in all samples. In all cases, the grains were poorly sorted and subangular to subrounded in shape. Moisture content was very low, which probably is due to the excessively drained nature of the soils. Also of interest was the absence of HCL reaction in Stratum IVb, which contains abundant marine shell. The HCL solution was applied to the quartz fraction, avoiding contact with shell fragments that contain calcium carbonate. The absence of any reaction to the HCL is unusual given the length of time since deposition of the shell. It suggests that that carbonate leaching into the surrounding sediments from the shell has not been extensive or that carbonates were rapidly leached out of the excessively drained sands.
Stratigraphic Interpretation Operation A and parts of Operation C are covered with clay fill to varying depths, with the shallowest occurrence in the open field in the northwest corner of the site and the deepest to the southeast of Building 591. The clay fill appears to represent a fill event that was designed to flatten out the topography. The fact that it appears in test units placed within the footprints of Buildings 588 and 591 indicates that it pre‐dates the construction of these buildings, which were built for the prison camp during the early 1970s (Campbell et al. 2009:68). Numerous historic features (large square postmolds, trash pits, refuse deposits, utility lines) encountered underneath the fill are related to military use of the property during the 1940s and 1950s. Historic features also have been identified by PTA in various locations within 8OK898 (e.g., Bourgeois et al. 2009; Campbell, Morehead et al. 2009).
119
Table 4.2. Characteristics of soil samples from TP 60. Wet Dry Moisture HCL FS# Color (wet) Description Wt. (g) Wt. (g) Content (%) Reaction Fine to Medium grained quartz sand with occasional Coarse quartz grains and charcoal, poorly sorted, 59 III 13 10YR4/3, brown 81.1 78 3.8% None quartz subangular to subrounded Fine to Medium grained quartz sand, with occasional Coarse quartz grains and shell fragments, rare 10YR3/2, very dark charcoal fragments, poorly sorted, quartz subangular 60 IVb 25 grayish‐brown 70.3 68.3 2.8% None to subrounded Fine to Coarse grained quartz sand with occasional 2.5Y3/2, very dark Very Coarse quartz, poorly sorted, subangular to 61 IVa 36 grayish‐brown 85 82.7 2.7% None subrounded Fine to Medium grained quartz sand, with occasional Coarse to Very Coarse grains, poorly sorted, 62 V 53 10YR4/3, brown 83.8 82.2 1.9% None subangular to subrounded 10YR6/4, yellowish‐ Fine to Coarse quartz sand, occasional charcoal, 63 VI 86 brown 90.4 89.2 1.3% None poorly sorted, quartz subangular to subrounded a Centimeters below unit datum (southwest corner). Depth Stratum (cmbud)a
Stratification and Chronology
The isolated fill deposits that were encountered throughout Operations A, B, and C are related to specific construction events or the installation of utilities. Many of these post‐ date the deposition of the clay fill as they penetrate into or through the clay. Meyer et al. 2002:43) indicate that as many as 7 fill episodes have been identified ranging from a few centimeters too as much as a meter in thickness. A major fill event was identified by PTA in the northern portion of Operation A within Buildings 632, 633, and 634. Excavations near the buildings encountered deep deposits of sand fill ranging from 65 to 95 cm in thickness (Campbell et al. 2009:77‐83). Several historic refuse pits, round postholes, and an asphalt slab were encountered. Deep fill deposits extend into Operation C where SEARCH exposed sand fill extending to about 95 cmbs. Fill in Operation B is more variable and appears to represent more recent activities. The fill overlies what PTA has interpreted as a buried plow zone (Ap) and which we designated as Stratum II (see Table 4.1). Because it was not frequently identified during our investigations, Stratum II is difficult to interpret. The interpretation of this stratum by PTA as buried topsoil or ground surface is plausible and is not contradicted by our field observations. Stratum III has been interpreted by PTA as a transitional A/C horizon consisting of sand that accumulated over and buried the midden after prehistoric abandonment of the site (e.g., Campbell et al. 1998:48; Meyer et al. 2002:45). Campbell et al. (1998:48) suggest that our Strata II and III reflect a “redevelopment of the Foxworth sand above the midden”; i.e., a new period of pedogenesis and the development of a new soil within the parent C material. There are several reasons to question this interpretation. Although Campbell et al. (1998) do not mention what the source of Stratum III sediments might have been, the NRCS indicates that Foxworth Series soils formed in sandy marine or eolian sediments. Deposition of marine sediments could have occurred as a result of higher seas levels or a major storm surge, but there is no evidence for sea levels higher than 1 m after 3000 BP (see Chapter 2) and the highest elevation of the midden (Stratum IV) is 3.61 m. A storm surge of that magnitude also is unlikely since the mainland is protected by barrier islands. Storm surges from hurricanes have overtopped these islands in the past, and the period from 3400 to 1100 BP was apparently quite active for hurricanes (Lui and Fearn 2000). Prehistoric hurricane overwash deposits at Western Lake near Grayton Beach range in thickness from a few millimeters to over 10 cm, but most are only a few centimeters thick; some contain small shell fragments (Liu and Fearn 2000:241‐242). Stratum III sands are as much as 45 cm thick in some locations and, except for fragments of large marine shells displaced from the underlying midden, no shells of small marine invertebrates are present. Multiple storm deposits are unlikely since there is no evidence of buried A horizons within the Stratum III sands which would have formed between storm events. Alternatively, Stratum III may be eolian in origin. However, the last period of substantial eolian deposition in the southeastern Coastal Plain was likely during the Hypsithermal, a
121
Bayou Park, 8OK898
period of dry climate, reduced flooding of interior rivers, and lowered lake levels which generally is considered to have occurred between about 9000 to 5000 years ago (Delcourt and Delcourt 1993; Holloway 2002:211‐226; Otvos 2004b; Stout and Spackman 2002:227‐ 235; Watts 1971), or before Bayou Park was being occupied by Elliott’s Point people. Some researchers (e.g., Leigh 1998, 2001) would argue that no significant eolian deposition has occurred in the Coastal Plain since the end of the Pleistocene. After about 4000 BP, rainfall increased, the rate of sea level rise slowed, and modern rainfall and vegetation regimes became established (Fearn and Cohen 1984:430; Watts and Hansen 1988). With the region heavily wooded, the landscape presumably stabilized and wind erosion decreased. These conditions would seem to make it unlikely that significant eolian deposition would have occurred after the Bayou Park midden was abandoned, which appears to have happened no earlier than the Mississippi period (AD 1000‐1500), although the exact date of abandonment is not known. On the other hand, Johnson and Fredlund (1993:71) argue that broad expanses of open pine forest to the northwest of Bayou Park would have contained enough open areas beneath the pine canopy to allow wind to pass through creating the conditions necessary for erosion of the unconsolidated surface sands and their subsequent redeposition at downwind locations. In support of this hypothesis, they point to a prehistoric feature at 8OK387 about 22.5 km northwest of Bayou Park, which was buried beneath 65 cm of eolian sand and was radiocarbon dated at 860 +120, ‐ 150 BP. An alternative explanation is that Stratum III represents a fill event, perhaps coeval with the deep fill event identified in Buildings 632, 633, and 634 and in TPs 95, 96, and 97. A marine shell from Stratum III was radiocarbon dated at 4140 ± 30 BP (cal 2381‐2141 cal BC). This is one of the older dates obtained from the site and it is the only date that is out of stratigraphic sequence, i.e., it was obtained from a higher elevation (18 cm below unit datum in TP60) than its date would suggest. This suggests that it was displaced from its original context and became mixed with Stratum III when it was being deposited. In several other locations modern rubble appeared to be mixed with what was identified in the field as Stratum III sediments. It is also quite possible that Stratum III represents a combination of eolian sands and sand fill, with the field identification of these depositional events made difficult by modern land modifications. Stratum IV is the prehistoric midden deposit that is present only in the northwestern portion of Operation A. PTA distinguishes between the sand midden (Ab1) and the shell midden (Ab2) while the original ground surface on which the midden accumulated is designated Ab3 (see Table 4.1). These correspond to our Strata IVa, IVb, and V, respectively. As discussed above, the shell deposits do not represent a distinct stratum or a buried surface (Ab), but instead are discontiguous lenses and deposits within the larger midden stratum. In addition, we identified a light‐colored, charcoal impregnated extension along the eastern periphery of the midden. 122
Stratification and Chronology
Campbell et al. (1998:48) indicated that Stratum V represents the original ground surface prior to midden formation and assigned this an Ab (buried A horizon) designation. This is probably true, but the color of Stratum V is the result of the leaching of organics and minerals from the overlying midden into the underlying yellowish‐brown sediments of the preexisting Stratum VI, the C horizon of the Foxworth Series. The fact that Stratum V is visible and most strongly expressed only where Stratum IV is present tends to support this interpretation.
SITE CHRONOLOGY
Radiocarbon Dates Twelve samples (8 charred wood, 4 marine shell) were submitted to Beta Analytic, Inc. for AMS dating. All of the marine shell samples were quahog clam (Mercenaria spp.). The results are presented in Table 4.3. Also included in this table are the four dates on marine shell (species unknown) obtained by PTA. All of the conventional radiocarbon ages have been calibrated using CALIB 6.01 (Hammer et al. 2001). Figure 4.10 is a plot of the calibrated ages and Figure 4.11 shows the calibrated radiocarbon dates. The charcoal sample from Feature 38, a large wood post, returned the youngest age at 250 ± 30 BP. The intercept crosses the calibration curve in several locations ranging from modern (‐1 to 11 cal BP) to between 428 and 376 cal BP (cal AD 1522‐1574). There is a 59% probability that the true age is between 324 and 270 cal BP, or cal AD 1620‐1680, and a 75% probability that the true age is between 428 and 270 cal BP, or cal AD 1522‐1680. However, no definitive contact‐period artifacts have been recovered from the site despite the numerous investigations that have been carried out over the past 20 years. While the probability that the sample is modern is only 3%, it remains a possibility and so interpretations of a contact‐period occupation of the site should be viewed with caution. Feature 51, another large post, returned a radiocarbon age of 1990 ± 30, or 1997‐1878 cal BP (48 BC‐AD 72). This would situate the feature within the Late Early Woodland period (see Chapter 2). The local prehistoric culture would have been Deptford, although no definitive Deptford artifacts have been recovered from the site. Feature 51 also closely resembles Feature 38, which dates much later in time, and both are at approximately the same elevation (3.20 m and 3.27 m amsl, respectively). Because both features provide dates for which there are no supporting artifacts, their usefulness in establishing the timing of post‐Archaic occupations is limited. The bulk of the radiocarbon dates are related to the Late Archaic period and these are extremely consistent. The maximum two‐sigma range of this occupation is 4481‐3766 cal BP (2529‐1817 cal BC), a period of about seven centuries. Using only the median dates calculated by CALIB 6.01 yields a minimum range of 4301‐3971 cal BP (2352‐2002 cal BC), or a little over three centuries.
123
Lab No.
Provenience
Material
Table 4.3. Radiocarbon dates from Bayou Park. Measured Age 13C/12C Conventional Calibrated (rcy BP) (‰) Age Age (2σ)a 270 ± 30 BP ‐26.1 250 ± 30 BP 428‐376 (.15) 366‐363 (
Lihat lebih banyak...
Comentários