Stone Tool Analysis in A Digital Environment:

1
Stone Tool Analysis in A Digital Environment

• Digital Imaging applied to a trampling experiment
  of edge and surface attrition in obsidian
• Authors E.S. Lohse, C. Schou, and D. Sammons

Paper presented at the Society for American
Archaeology Meetings, 2005
2
Lit-review specific to trampling

• Issues on recognizing trampling attrition vs.
  patterned use wear
• Flenniken and Haggerty 1979
• Gifford-Gonzalez et al. 1985
• Pryor 1988
• Nielsen 1991
• McBrearty et al. 1998

• Ideas on character of trampling attrition
• Shea and Klenck 1993
• McBrearty et al. 1998
• Damage from agricultural equipment
• Prost 1988
• Mallouf 1981

3
Research Context

• Key to effective interpretation of lithic
  materials in archaeology is consistent
  discrimination of unintentional attrition from
  patterned attrition resulting from deliberate
  manufacture and use.
• Digital imaging allows accurate measurement of
  micropatterns important in defining distinctive
  patterns of attrition.
• A trampling experiment was designed that
  controlled for forces of compression, variability
  in raw materials, and depositional environment.

4
Trampling Study Research Hypothesis

• Hypothesis (Ho) There will be no significant
  difference in edge damage and surface attrition
  between thick blocky primary obsidian conchoidal
  flakes and small, thin tertiary obsidian
  conchoidal flakes in a constant depositional
  environment exposed to controlled trampling.

5
Trampling Study Forms
Primary flake
Secondary flake
Tertiary flake
Bladelet
6
Concept Map Attrition Potential
7
Trampling Study Parameters

• Study group
• N67 conchoidal flakes removed from a single
  Glass Buttes obsidian core through free-hand
  percussion flakes arranged in four sampling
  strata
• 1. Blocky, thick primary removals
• 2. Thinner, secondary removals
• 3. Smaller, thin symmetrical tertiary removals
• 4. Small thin symmetrical blades

• Depositional environment
• All N67 flakes were placed in a trampling box
• 1. Box dimensions 2X4 in outline and 6"
  in depth
• 2. Box strata
• a. Plywood base ¾"
• b. Clay loam bed 2"
• c. Fine earth layer 2"
• d. Fine sand layer 2"

8
Trampling Study Methodology

• Methodology
• 1.  Unmodified flakes were collected as removed
  from the core
• 2.  Unmodified flakes were photographed with a
  digital camera at 300 dpi resolution and saved at
  2" maximum dimensionVentral and Dorsal views
• 3.   Unmodified flakes were randomly scattered in
  the trampling box and raked below the surface
  with analysts' fingers
• 4.   The trampling box was placed at the door of
  the classroom and students were asked to walk
  through the box in and out of the classroom
• 5.   At the end of one week (3 classes N25
  tramplers two trips) the specimens were removed
  from the box
• 6.   Trampling and recording procedures were
  repeated over the course of three weeks.

9
Surface Plots

• All specimens were recorded as unmodified digital
  images and as surface plots recording pixel
  values
• Surface plots are three-dimensional
  representations of the intensity of an image
• X length
• Y width
• Z height
• Viewpoint elevation provides perspective from
  0-90 degrees rotation moves the viewer around
  the object in a 180 degree arc
• Surface plots could then be reduced to more
  simplified images using a range of filters and
  transforms to derive accurate edge boundaries

10
Need to Simplify Image for Data Recording

• To simplify recording, a practical schema was
  applied
• This schema allowed analysts to delineate
  measurement areas, greatly reducing areal
  coverage requirements
• Categories relate to projected differences in
  potential attrition relative to structural
  characteristics of the flake

11
Recording Random Attrition Proportional Areas
12
Recording Patterned Attrition Proportional Areas
13
Measurement in Pixel Environment
14
Real Images to Surface Plots
Left, image of an unmodified unused obsidian
flake, 300 dpi, Canon video camera with lenses
and fiber optic light source Right, surface plot
of same image at 45 degree orientation using a
shaded/illuminated style. Image-Pro Plus software.
15
AOIs and Edge Rotations
Upper left, digital image of obsidian flake 1-1
upper right, aoi of unworn wedge subsequent to
modification lower right, surface plot of unworn
edge at 45 degrees. Lower left, same surface plot
rotated at 180 degrees with viewer perspective
from the aoi midline margin toward the edge.
16
Perspective in Reading the Plots
These are oblique views of the same obsidian tool
1-1 showing how different Perspective and light
change definition of edges and surfaces. Upper
left, flaked tool edge at X100 Scraping strokes.
Lower left, tool edge at X300 strokes. Right,
surface plots.
17
Trampling Study (1-3 weeks)
Specimen 1-09, dorsal view, 1-3 weeks Trampling.
Upper left, fresh flake upper center, surface
plot-fresh surface at 45 rotation upper right,
surface plotfresh surface at 90 rotation lower
left, flake after 3 trials lower right, surface
plot- surface at 3 weeks at 90 rotation.
18
Results No significant attrition over three
trials

• Thin edges and surfaces showed no measurable
  attrition
• Why?
• Infer that sand matrix (grain size and texture)
  cushioned and distributed weight of foot traffic,
  allowing fragile flake edges and surfaces to
  rotate and still be uniformly supported
• Next step manipulate sand matrix, including more
  variable grain size and texture
• Goal understand breakage by relating attrition
  characteristics to variable grain size and grain
  texture

19
Conclusions

• Trampling Exercise and Archaeological Context
  Problems with sand
• Digital Imaging and Analysis

20
Problems with the sand

• It appears that the sand was too fine, very
  regular in size and surface texture
• This allowed obsidian flakes to be evenly
  supported by the sand as compressive forces moved
  the flakes rotationally through the matrix

21
Overview Sand

• Robertson (1990) CPT soil classification scheme.
• 1 Sensitive, fine grained2 Organic
  soils-peats3 Clays-clay to silty clay4 Silt
  mixtures-clayey silt to silty clay5 Sands
  mixtures-silty sand to sandy silts6
  Sands-clean sand to silty sand 7 Gravelly sand
  to sand8 Very stiff sand to clayey sand9
  Very stiff, fine grained
• Heavily overconsolidated or cemented

• The United States Golf Association considers
  seven factors when selecting bunker sand
  particle size, particle shape and penetrometer
  value, crusting potential, chemical reaction and
  hardness, infiltration rate, color, and overall
  playing quality. Depending upon your location and
  climate, how you rank these factors may vary
  slightly.
• The biggest factor, the fried egg test, or in
  testing terminology, the Penetrometer value. The
  penetrometer value measures the energy required
  to bury a ball in sand. This value shows the
  ability of sand to resist the golf ball from
  burying, or in more scientific terms, its
  resistance to compression.

http//cgiss.boisestate.edu/billc/SAGEEP98/Tab1.h
tmlTab1 http//www.findarticles.com/p/articles/mi
_qa4031/is_200301/ai_n9171064
22
Sand Mix and Penetration
Beard, James B., Turf Management for Golf
Courses, 2002. http//www.turfdiag.com/bunker.htm
_
23
Measuring Sand Characteristics

• Ball-lie Rating Penetrometer Value (kg/cm2)
  Rating
• Highlt 1.8 Undesirable
• Moderate1.8 to 2.2 Acceptable
• Slight2.2 to 2.4 Acceptable
• Very Lowgt 2.4 Desirable
• Shape-Crusting or Set-Up Rating
• Rounded Severe Undesirable
• Sub-Rounded to Mixed Slight to Moderate
  Acceptable
• Angular None Desirable

http//www.turf-tec.com/PN5lit.html
Evaluation process 1. Particle size analysis
- ASA methods of soil analysis - ASTM
methods 2. Infiltration rate testing 3.
Evaluation - research questions -
test parameters
http//www.turfdiag.com/bunker.htm_
24
USGA Sand Sorting
http//www.turfdiag.com/sand_technology.htm http/
/www.rickly.com/sai/VASTA-01.htm
25
Potential for Digital Imaging

• Digital imaging allows automated recording of
  complex shapes in a controlled environment
  (scale, lighting, measurement)
• Measurement will be accurate to the level of a
  single pixel value and allows recording of
  complex shapes
• Measurement in a digital environment will allow
  creation of ratios and proportional area
  measurements applicable to a range of research
  questions in lithic analysis in archaeology

26
Archaeological Conclusions

• McBrearty et al. 1998 published these
  conclusions
• Found a high degree of damage or edge
  modification to artifacts trampled in
  fine-grained sediments (cf. Flenniken and
  Haggerty 1979)
• Most damage found when artifacts were trampled on
  loam artifact to artifact damage
• Damage related to penetrability of matrix (see
  also Nielsen 1991)
• Trampling attrition should be positively
  correlated with artifact densities

• This study
• Finely sorted and rounded sand grains inhibit
  trampling attrition by supporting the artifacts
  in all rotational positions
• Attrition should be positively correlated with
  inducing variability in sand grain size and
  different sand textures
• Sandy matrices will probably inhibit attrition
  even in the case of high density artifact
  accumulations

27
Modification Trajectory Steps 2-5
2 bifacial pressure flaking
5 scraping at X300
3 further bifacial reduction
4 scraping at X100
28
To Do Induce Greater Attrition

• Rebuild the trampling experiment to induce
  greater attrition
• Create sand mix to emphasize greater size
  irregularity and angular surface texture
• Create thinner sand layer underlain by earth or
  loam layer
• Introduce more obsidian flakes and or reduce the
  trampling area to increase densities

• Employ digital imaging schema developed here to
  record attrition on obsidian specimens
• With enhanced attrition, use protocols to create
  accurate measurements
• Ratios
• Proportional areas
• Create statistical correlations between material
  and flake types and staged exposure to
  compression in variable sand matrices

29
Suggested References

• Beard, James B., Turf Management for Golf
  Courses, 2nd Ed. p. 259-281.  Ann Arbor Press, 
  Chelsea, MI. 2002. See Turf Diagnostics Design
  Helping You Have Healthy Turf, http//www.turfdiag
  .com/bunker.htm_
• Blott, Simon J., Ali M. Al-Dousari, Kenneth Pye
  and Samantha E. Saye, Three-Dimensional
  Characterization of Sand Grain Shape and Surface
  Texture Using a Nitrogen Gas Adsorption
  Technique,, Journal of Sedimentary Research
  January 2004 v. 74 no. 1 p. 156-159.
  http//jsedres.geoscienceworld.org/cgi/content/ful
  l/74/1/156TB1
• Brown, K. W. and Thomas, J.C.  1986.  Bunker
  Sand Selection.  Golf Course Management. 
  5464-70.
• Clement, W. P., Cardimona, S., and
  Kadinsky-Cade, K., 1997a, "Geophysical and
  geotechnical site characterization data at the
  Groundwater Remediation Field Laboratory, Dover
  Air Force Base, Dover, Delaware," Proc. SAGEEP,
  pp. 665-673.

30
References. Continued.

• Endres, Anthony L. and William P. Clement,
  Relating Cone Penetrometer Test Information to
  Geophysical Data A Case Study.
  http//cgiss.boisestate.edu/billc/SAGEEP98/CPT.ht
  ml
• Flenniken, J. and and J. Haggerty. 1979.
  Trampling as an Agency in the Formation of Edge
  Damage An Experiment in Lithic Technology.
  Northwest Anthropological Research Notes 13
  208-14.
• Gifford-Gonzalez, D.D. et al., 1985, The third
  dimension in site structure an experiment in
  trampling and vertical displacement. American
  Antiquity 50 803-818.
• Klute, A. (ed.), Hydraulic Conductivity of
  Saturated Soils.  1986.  Methods of Soil Analysis
  Vol. 1, Agronomy 9687-703.  Amer. Soc. of
  Agronomy, Madison,  WI.
• Mallouf, Robert J., 1981, A Case Study of Plow
  Damage to Chert Artifacts the Brookeen Creek
  Cache, Hill County, Texas. Texas Historical
  Commission, Office of the State Archaeologist,
  Report 33, Austin.

31
References. Continued.

• McBrearty, Sally et al., 1998, Tools underfoot
  human trampling as an agent of lithic artifact
  edge modification, American Antiquity 63
  108-129.
• Nielsen, A. E., 1991, Trampling the
  Archaeological Record An Experimental Study.
  American Antiquity 56(3) 483-503.
• Prost, Dominique, 1988, Essai detude sur les
  mecanismes denlevement produits par les facons
  agricoles et le pietinement humain sur les silex
  experimentaux. In Industries lithiques
  Traceologie et technologie, S. Beyries (ed.), pp.
  49-63. BAR International Series 411, Oxford.
• Pryor, John H., 1988, The effects of human
  trample damage on lithics a consideration of
  crucial variables. Lithic Technology 17 45-50.

32
References. Continued.

• Shea, J.J. and J.D. Klenck, 1993, An
  experimental investigation of the effects of
  trampling on the results of lithic microwear
  analysis, Journal of Archaeological Science 20
  175-194.
• Zhang, Z., and Tumay, M. T., 1996,
  "Simplification of soil classification charts
  derived from the cone penetration test,"
  Geotechnical Testing Journal, v. 19, pp. 203-216.