ArticlePDF Available

Reconstruction of woodland vegetation and firewood exploitation in Nine Mile Canyon, Utah, based on charcoal and pollen analysis


Abstract and Figures

Anthracological (charcoal) and pollen analysis conducted on samples from multiple sites in Nine Mile Canyon, Utah, western United States of America, were used to reconstruct the woodlands of this region and interpret firewood exploitation during the Formative period (∼ AD 200–1300). The pollen record identifies constituents of the paleoenvironment of this region, reflecting species that grew in the vicinity of the site and in the broader area. Wood charcoal reflects various trees and shrubs that were used as fuel or for construction by the prehistoric occupants settled in Nine Mile Canyon. Pollen results elucidate a portion of the spectrum of wood taxa, which varies in different parts of the canyon, depending on elevation and the canyon's topography. While micro-charcoal analysis has a long and rich tradition in association with pollen analysis, anthracological analysis is different in that it focuses on macroscopic pieces of archaeological charcoal. Here, anthracology, using SEM imagery, helps define local woody taxa by identifying macroscopic pieces of archaeological charcoal. These two analytical tools build a complementary record of local woodland vegetation through aerial pollen transport and fuel wood selection. While the pollen record is affected by aerial transport, the anthracological assemblages, in most cases, reflect species growing in the vicinity of the prehistoric settlements, but can be affected by natural transport within the drainage. Recovery of Douglas-fir (Pseudotsuga menziesii) charcoal from the lower portions of the canyon, containing steep walls and natural conditions not favorable for Douglas-fir growth, indicates presence of drift-wood or debris-flow deposits.
Content may be subject to copyright.
Reconstruction of woodland vegetation and rewood exploitation in
Nine Mile Canyon, Utah, based on charcoal and pollen analysis
Peter Kov
, Linda Scott Cummings
PaleoResearch Institute, 2675 Youngeld Street, Golden, CO, 80401, USA
article info
Article history:
Received 13 July 2016
Received in revised form
10 May 2017
Accepted 21 June 2017
Available online 5 July 2017
Charcoal analysis
Pollen analysis
Spatial analysis
Woodland vegetation reconstruction
Nine Mile Canyon
Anthracological (charcoal) and pollen analysis conducted on samples from multiple sites in Nine Mile
Canyon, Utah, western United States of America, were used to reconstruct the woodlands of this region
and interpret rewood exploitation during the Formative period (~ AD 200e1300). The pollen record
identies constituents of the paleoenvironment of this region, reecting species that grew in the vicinity
of the site and in the broader area. Wood charcoal reects various trees and shrubs that were used as fuel
or for construction by the prehistoric occupants settled in Nine Mile Canyon. Pollen results elucidate a
portion of the spectrum of wood taxa, which varies in different parts of the canyon, depending on
elevation and the canyon's topography. While micro-charcoal analysis has a long and rich tradition in
association with pollen analysis, anthracological analysis is different in that it focuses on macroscopic
pieces of archaeological charcoal. Here, anthracology, using SEM imagery, helps dene local woody taxa
by identifying macroscopic pieces of archaeological charcoal. These two analytical tools build a com-
plementary record of local woodland vegetation through aerial pollen transport and fuel wood selection.
While the pollen record is affected by aerial transport, the anthracological assemblages, in most cases,
reect species growing in the vicinity of the prehistoric settlements, but can be affected by natural
transport within the drainage. Recovery of Douglas-r(Pseudotsuga menziesii) charcoal from the lower
portions of the canyon, containing steep walls and natural conditions not favorable for Douglas-r
growth, indicates presence of drift-wood or debris-ow deposits.
©2017 Elsevier Ltd and INQUA. All rights reserved.
1. Introduction
Although wood charcoal is one of the most common, and
many times the only, remains recovered from archaeological sites
in the western United States, its interpretative value often has
been limited to AMS radiocarbon analysis. Archaeobotanical
research usually focuses on identication and analysis of food
related plants (corn, cucurbits, beans, etc.) and utilization of
seeds, fruits, and nuts from wild plants. Paleoenvironmental and
paleoclimatic reconstructions, as well as re history (Whitlock
and Larsen, 2001) are conducted using palynology and dendro-
chronology (Baillie, 1982), and lately also phytolith analysis
(Blinnikov, 2005). Systematic anthracological studies (Asouti,
2003, 2013; Asouti and Austin, 2005; Asouti et al., 2015; Elliott,
2012; Salavert and Dufraisse, 2014; Scheel-Ybert, 2001; Wright
et al., 2015, 2017) provide proof that wood charcoal analysis is a
powerful tool for prehistoric vegetation reconstruction and un-
derstanding woodland modication and change, as well as
ancient fuel management. Studies that combine anthracological
results with pollen, phytolith, carpological, and dendrochrono-
logical analysis provide even better understanding of the rela-
tionship between prehistoric societies and their environment. In
locations where long stratigraphic lake sediment and/or bog re-
cords are available macro-charcoal assemblages from archaeo-
logical sites may be compared to relevant regional palynological
records obtained from radiocarbon-dated cores (Emery-Barbier
and Thi
ebault, 2005; N
adi et al., 2012; Nelle et al., 2010;
Nocus et al., 2011; Novak et al., 2011).
This paper aims to present the anthracological investigation of
prehistoric features excavated along Nine Mile Creek in Nine Mile
Canyon, east-central Utah as an environmental proxy. Dense
occupation of the canyon during the Formative period (~AD
200e1300) allows spatial examination of woodland composition in
relation to topographical settings of the canyon. Conjunction of
*Corresponding author.
E-mail addresses:,
(P. Kov
cik), (L. Scott Cummings).
Contents lists available at ScienceDirect
Quaternary International
journal homepage:
1040-6182/©2017 Elsevier Ltd and INQUA. All rights reserved.
Quaternary International 463 (2018) 312e326
these data with pollen analysis provides an ideal tool for micro-
regional woodland reconstruction and ancient rewood manage-
ment on the West Tavaputs Plateau.
2. Regional setting
2.1. Geographic and environmental setting
Situated at the northern edge of the West Tavaputs Plateau
N longitude), at the border of Carbon and Duchesne
Counties in east-central Utah, Nine Mile Canyon is one of the
most signicant areas associated with the Fremont archaeological
complex in the northern portion of the Colorado Plateau (Fig. 1).
The west-east trending canyon, with the perennial Nine Mile
Creek, drains into Green River in Desolation Canyon. Elevation of
the canyon bottom ranges from approximately 2300 m asl
(7600 ft) in the west to 1400 m asl (4600 ft) at the conuence of
Nine Mile Creek with Green River in the east. Walls of the canyon
often rise 150e300 m (~500e1000 ft) above the canyon oor,
while plateau peaks reach an altitude of approximately 3100 m
asl (10200 ft asl) (Bruin Point). High plateaus and rugged terrain
with deep canyons dominate the Tavaputs Plateau. In combina-
tion with a semi-arid mid-continental climate with limited, but
year-round precipitation (Bureau of Land Management, 2011;
Knight et al., 2010:108), the Tavaputs Plateau supports various
vegetation types. Pinyon-juniper (Pinus edulis-Juniperus spp., but
mostly J. osteosperma) woodlands and shrublands covering dry
and hot south- and east-facing canyon slopes (Fig. 2:A) represent
important plant communities for prehistoric people of this region
(Lanner, 1975:1, 1981, 1983). Edible pinyon nuts and juniper
berries were collected for food and medicinal purposes, while
wood was used for construction and fuel. Various grasses (Poa-
ceae), yucca (Yucca spp.), pricklypear cactus (Opuntia spp.),
jointr(Ephedra spp.), serviceberry (Amelanchier spp.), and
mountain mahogany (Cercocarpus spp.) are mixed within the
pinyon-juniper ecosystem. Bottoms of the slopes and dry plains
support sagebrush-greasewood (Artemisia spp.-Sarcobatus ver-
miculatus)shrublands(Fig. 2:B) (Mozingo, 1987), with rabbit-
brush (Chrysothamnus spp.), snakeweed (Gutierrezia spp.),
saltbush (Atriplex spp.), and Indian ricegrass (Achnatherum spp.).
Shrubby willows (Salix spp.), cottonwoods (Populus spp.), maples
(Acer spp.), and cattails (Typha spp.) grow along the creek
(Fig. 2:C). Higher elevations and northern exposures support
Douglas-r(Pseudotsuga menziesii)stands(Fig. 2:D), while high
altitude montane conifer forests consist of r(Abies spp.) and
spruce (Picea spp.). Today, the canyon bottom is heavily affected
by cattle ranching and natural gas wells, and a small agricultural
presence is noted.
2.2. Cultural context
The rst archaeological explorations of Nine Mile Canyon
quickly followed its discovery by Euroamericans in the second half
of the nineteenth century and throughout the twentieth century
(Gillin, 1938, 1955; Marwitt, 1986; Morss, 2009; Patterson, 2016;
Spangler, 2000a:25e38; 2000b, 2013; Spangler and Spangler,
2010). Fremont villages with pithouses on the canyons bottoms,
cliff dwellings and granaries (Fig. 3:A), circular structures (Fig. 3:B)
on rock outcrops and mesas, as well as countless rock art panels
(Fig. 3:C) evidence active life during the Formative Period associ-
ated with spread of farming to the Tavaputs Plateau (Patterson and
Flanigan, 2010; Patterson, 2015, 2016; Spangler, 2013:165). Agri-
cultural elds in the canyon oodplains and alluvial fans were used
to grow maize (Zea mays) and possibly other cultigens including
beans (Phaseolus spp.) and cucurbits (Cucurbitaceae). Previously,
major Fremont occupation of the canyon was presumed around AD
1000 (Spangler, 2013:165). Based on new radiocarbon evidence,
Patterson (2015, 2016) suggests a continuous occupation of Nine
Mile Canyon from at least AD 500 to AD 1250, extending to the
Early and Middle Formative periods. Recovery of corn kernels/
Fig. 1. Map of southwestern United States of America, indicating Nine Mile Canyon (short blue line) on Tavaputs Plateau (approximate boundaries indicated with black line) within
the Colorado Plateau (approximate boundaries indicated with white line) (compiled by P. Kov
cik). (For interpretation of the references to colour in this gure legend, the reader is
referred to the web version of this article.)
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326 313
cupules and corn pollen (Kov
cik et al., 2014; Puseman and Scott
Cummings, 2013) from Early and Middle Formative features sug-
gests corn utilization in the canyon prior to AD 950.
The most recent archaeological investigation in the Nine Mile
Canyon is associated with discovery of natural gas deposits on the
plateau, and subsequent infrastructure development in the canyon
in the 2000s. Several sites, buried under the canyon's alluvial de-
posits, were excavated during the pipeline and road reconstruction
in 2012e2013 by Montgomery Archaeological Consultants under
the directorship of Jody Patterson. The sites varied from a single
feature to complex multifunctional settlements. Hearths, roasting
pits, storage pits, and pithouses from 27 sites were sampled for
macrooral and pollen analysis to investigate the prehistoric
environment and plant use by canyon's prehistoric occupants. Re-
covery of charred Zea mays cobs, cupules, kernels, and pollen
indicate intensive corn agriculture. In addition, wild plant types,
including goosefoot (Chenopodium), sagebrush (Artemisia), care-
lessweed (Cyclachaena xanthifolia), tansy mustard (Descurainia),
pepperweed (Lepidium), pigweed (Amaranthus), saltbush (Atriplex),
cattail (Typha), pine/twoneedle pinyon (Pinus/Pinus edulis), and
hedgehog cactus (Echinocereus) also were utilized (Kov
cik et al.,
2014; Puseman and Scott Cummings, 2013).
3. Material and methods
3.1. Sampling
Samples were collected from 67 features at 27 sites (Fig. 4) along
Nine Mile Creek in the upper and middle portions of Nine Mile
Canyon (approximately 37 km). The examined sites more or less
clustered into six zones, each unique in elevation range and canyon
wall steepness. Designated Zones I through VI, following west-east
canyon trending, decrease in elevation and increase in slope
In most cases, both macrooral (250e700 ml) and pollen (up to
50 ml) samples were collected from the same features ensuring
chronological compatibility of the data. Sampled features include
pit structures, storage pits, roasting pits, and hearths buried under
alluvial fans along the Nine Mile Creek. The majority of features/
sites are buried three to four, or up to six meters below the modern
ground surface (Patterson, 2016) as shown in Fig. 5.
Pollen grains, deposited during feature use and/or immedi-
ately after its abandonment, indicate species growing in the vi-
cinity of the site as well as in the broader region, identifying the
environmental potential of the study area during the occupation.
As the features were buried, their pollen records, which accu-
mulated during use and immediately post use, were sealed,
Fig. 2. Four major plant environments observed in Nine Mile Canyon include pinyon-juniper woodland (A), sagebrush-greasewood oodplain shrubland (B), riparian zone (C), and
montane forest with Douglas-r stands (D) (photo by P. Kov
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326314
providing accurate representation of use and local vegetation for
each occupational period. Frequent high-energy surface runoff
events in the canyon naturally sealed many of the prehistoric
Charcoal is present in feature ll as a result of human activity,
the most common of which involved burning wood as fuel. In
accordance with the Principle of Least Effort,thecharcoal
assemblages directly reect species availability, as well as effort-
less acquisition of fuel wood (Shackleton and Prins, 1992). How-
ever, several studies (Dussol et al., 2016; Henry and Th
2014; Marston, 2009; Picornell-Gelabert et al., 2011) suggest that
socio-economic strategies and human preference also have an
effect on the anthracological record, thus they must be considered.
3.2. Anthracological methods
Macrooral samples were oated using a modication of pro-
cedures outlined by Matthews (1979). Each sample was added to
approximately three gallons of water, then stirred until a strong
vortex formed. The oating material (light fraction) was poured
through a 250-
m mesh sieve. Additional water was added and the
process repeated until all oating material was removed from the
samples (a minimum of ve times). The material remaining in the
bottom (heavy fraction) was poured through a 0.5-mm mesh
screen. The oated portions and heavy fractions were allowed to
The light fractions were weighed, then passed through a series
of graduated screens (US Standard Sieves with 4-mm, 2-mm, 1-
mm, 0.5-mm, and 0.25-mm openings) to separate charcoal
debris and to initially sort the remains. Charcoal pieces larger than
2 mm, 1 mm, and 0.5 mm in diameter were separated from the
smallest sized particles, and the total charcoal was weighed. At
least 40 charcoal fragments were identied from each sample.
Although examining a minimum of 20 pieces of charcoal has been
set as a standard in the western United States (Minnis, 1978:9;
Minnis and Raymer, 1982:2), we prefer to examine and identify a
minimum of 40 pieces of charcoal, when they are present. Often
charcoal fragments larger than 2-mm are not abundant and
smaller pieces (1-mm and 0.5-mm) must be examined to meet
this minimum number of fragments. In cases where charcoal
fragments larger than 2 mm were absent or insufcient (less than
40 pieces), fragments smaller than 2 mm were examined, starting
with fragments larger than 1 mm. However, a few samples did not
yield even 40 identiable charcoal fragments, excluding uniden-
tied hardwood and conifer. Charcoal pieces in a representative
sample were broken to expose fresh transverse, radial, and
tangential sections, then examined under a binocular microscope
at a magnication of 70and under a Nikon Optiphot 66 micro-
scope at magnications of 320e800x. Weights of each charcoal
type within the representative sample were recorded. Charcoal
remains were identied by comparison with modern and
archaeological references curated at PaleoResearch Institute (PRI)
in Golden, Colorado (USA) and using identication manuals
(Carlquist, 2001; Hoadley, 1990; Minnis, 1987; Panshin and de
Zeeuw, 1980; Schweingruber et al., 2011, 2013). Images were ob-
tained using a PhenomWorld desktop SEM (scanning electron
3.3. Palynological methods
Sediments present unique challenges for pollen preservation
and recovery. In general, the western United States has alkaline
sediments and limited rainfall, providing better pollen preservation
than sediments in areas that experience greater rainfall. A chemical
extraction technique based on otation is the standard preparation
technique used in this laboratory for recovering pollen grains from
sediments. This particular process was developed for extracting
pollen from soils where the preservation has been less than ideal
and the pollen density is lower than in peat. Hydrochloric acid
(10%) was used to remove calcium carbonates present in the sedi-
ment samples, after which, they were screened through 250-
mesh. Multiple water rinses until neutral employ Stoke's Law for
Fig. 3. Cliff dwellings and granaries (A), circular structures situated on the rock out-
crops (B), and abundant rock art panels (C) indicate presence of Fremont occupants in
Nine Mile Canyon (photo by P. Kov
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326 315
settling time. After settling the supernatant was poured off. A small
quantity of sodium hexametaphosphate was mixed into each
sample to suspend clay-sized particles prior lling the beakers with
water. Again, multiple rinses employing Stoke's Law and decanting
facilitated clay removal. Treatment with sodium hexametaphos-
phate was repeated, as necessary, to remove clay. This process was
repeated with ethylenediaminetetraacetic acid (EDTA), which
removes clay, soluble organics, and iron. Finally, the samples were
freeze-dried under vacuum.
Once dry, the samples were mixed with sodium polytungstate
(SPT), at a density of 1.8 g/ml, and centrifuged to separate the
organic material including pollen and starch, which oats, from
the inorganic remains and silica, which do not oat. The super-
natant containing pollen and organic remains was decanted and
retained. The sodium polytungstate process was repeated to
recover all of the organics. Once the organics were recovered, the
accumulated supernatant was centrifuged at 1500 rpm for 10 min
to allow small-sized silica to be separated from the organics. This
supernatant was decanted into a 50-ml conical tube and diluted
with reverse osmosis deionized (RODI) water and centrifuged at
3000 rpm to concentrate the organic fraction in the bottom of the
tube. The pollen-rich organic fraction was rinsed, then all sam-
ples received a short, 25 min, treatment in hot hydrouoric acid
(49%) to remove remaining inorganic particles. The samples were
Fig. 4. Map of Nine Mile Canyon, Utah, indicating six zones and archaeological sites sampled for anthracological and pollen analysis (compiled by P. Kov
Fig. 5. Features/sites in Nine Mile Canyon are buried under several meters of alluvial fan deposits (black arrows indicate a buried hearth exposed in one of the arroyos) (photo by P.
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326316
acetylated (9 parts acetic anhydride and 1 part 95e95% sulfuric
acid) at an altitude of 1610 m (5280 ft) for 10 min to remove
extraneous organic matter. This process proceeds at a lower
temperature at this elevation, permitting longer exposure to the
acetolysis mixture. The samples were rinsed with RODI water to
neutral. A few drops of 5% potassium hydroxide (KOH) were
added to each sample, which was then stained lightly with
A Nikon binocular light microscope was used to count pollen at
a magnication of 500. Pollen preservation in these samples was
good with few grains so badly damaged that they were not
identiable. All counts were obtained on a single microscope slide
per sample. Percents were calculated for each sample using total
pollen representing only the woody plants observed in each
sample (listed on the pollen diagram). An extensive comparative
reference housed at PaleoResearch Institute aided pollen identi-
cation to the family, genus, and species level, where possible. Low-
spine Asteraceae are dened as having the morphology of Am-
brosia,Franseria,orXanthium, while High-spine Asteraceae are all
other Asteraceae.
3.4. Quantication, multivariate statistics, and spatial analysis
Identied taxa in the anthracological data set were quantied
applying absolute count (c), percentage abundance (%), and ubiq-
uity (u). Quantication of the pollen grains from arboreal and
shrubby species was undertaken using percentage abundance. The
percentage pollen diagram for arboreal and shrubby species was
produced using Tilia 1.7.16. Total pollen concentrations were
calculated in Tilia using the quantity of sample processed in cubic
centimeters (cc), the quantity of exotics (spores) added to the
sample, the quantity of exotics counted, and the total pollen
counted and expressed as pollen per cc of sediment. When multiple
samples from one site were combined into a single entry for the
pollen diagram the total pollen concentrations for those samples
were averaged. Principal components analysis (PCA) was
conducted on the anthracological record to test separation between
various plant communities in the canyon.
4. Results
Out of 27 pollen types identied in the pollen records, only
pollen grains representing six arboreal and nine shrubby woody
taxa (Table 1), which have the potential to be recovered in the
anthracological assemblages, were used for pollen percentage
abundance and statistical analyses. Pollen from grasses, forbs, and
culturally important taxa were excluded from this study to elimi-
nate bias introduced by the presence of weedy and food taxa.
Approximately 97% of 2637 analyzed charcoal fragments were
identied to 15 different taxa (Table 1). Unidentied hardwood and
conifer fragments either were too small or too vitried for further
identication. These fragments within the anthracological record
were insignicant, thus were excluded from the fragment count
and ubiquity calculations after Asouti (2003) and Wright et al.
(2015, 2017).
4.1. Pollen dataset
Regional pollen records suggest that the vegetation around
1000 CAL yr. BP does not signicantly differ from the modern
conditions (Klimaszewski-Patterson and Mensing, 2016;
Louderback and Rhode, 2009; Louderback et al., 2015). The pol-
len record from 44 features at 25 sites is ranked by pollen taxon
dominance (Table 1). Amaranthaceae (goosefoot/amaranth family
eformerly termed Cheno-am), Pinus (pine), Sarcobatus vermic-
ulatus (greasewood), Artemisia (sagebrush), and Asteraceae (sun-
ower family) pollen are generally represented at most of the sites
in the study area. Ephedra (jointr), Juniperus (juniper), Abies (r),
Quercus (oak), Pseudotsuga menziesii (Douglas-r), Rosaceae (rose
family), Rhus (sumac), and Picea (spruce) each represent a taxon
with a less than 1% of total pollen grain count. Under or over
representation of some taxa is associated with individual species
Table 1
Major plant communities indicated in the pollen and anthracological (charcoal) record from Nine Mile Canyon sites. In the pollen record, Ephedra (jointr) pollen includes
Ephedra nevadensis-type (Nevada jointr) and Ephedra torreyana-type (Torrey's jointr), while Asteraceae (sunower family) pollen includes Low- and High-spine Asteraceae.
Fragment count (C), percentage abundance (%), and ubiquity (U) for pollen and anthracological (charcoal) assemblages from examined sites (compiledbyP.Kov
Plant Community Taxon Totals
Pollen Charcoal
Scientic Name Common Name C % U C % U
Montane Abies Fir 24 0.38 40.00
Picea Spruce 1 0.02 4.00
Pseudotsuga menziesii Douglas-r 6 0.09 20.00 336 13.20 55.60
Pinyon-juniper Juniperus Juniper 31 0.49 48.00 359 14.10 63.00
Pinus/Pinus edulis Pine/Twoneedle pinyon 1825 28.60 100.00 535 20.90 66.70
Quercus Oak 23 0.36 48.00
Ephedra Jointr 63 0.99 76.00 5 0.20 3.70
Rosaceae Rose family 5 0.08 16.00 2 0.10 7.41
Amelanchier Serviceberry 59 2.30 29.60
Cercocarpus Mountain mahogany 1 0.04 3.70
Rhus Sumac 4 0.06 4.00 2 0.10 3.70
Floodplain Asteraceae Sunower family 409 6.40 100.00 11 0.40 11.10
Artemisia Sagebrush 685 10.70 100.00 290 11.40 55.60
Chrysothamnus Rabbitbrush 23 0.90 18.50
Sarcobatus vermiculatus Greasewood 1337 20.90 92.00 715 28.00 74.10
Amaranthaceae Amaranth and Goosefoot family 1976 30.90 100.00
Atriplex Saltbush 14 0.50 18.50
Acer Maple 2 0.10 7.41
Populus/Salix Cottonwood/Willow 201 7.90 25.90
Totals 6389 100.00 2555 100.00
# of Taxa 13 15
# of Sites 25 27
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326 317
pollen productivity and pollen transportability. Various pollen
percentages of different species within one feature or site (Fig. 6)
do not necessarily indicate abundance or lack of certain taxa in the
vicinity of the site. Pines, for instance, produce large quantities of
pollen grains transported over long distances, while Juniperus,
Rosaceae, and Rhus produce less pollen with shorter transport
distances. Total pollen concentrations were in the multiple hun-
dreds to more than one hundred thousand pollen per cc of sedi-
ment, indicating good pollen preservation.
4.2. Anthracological dataset
The anthracological assemblage, with 2555 fragments identied
in 66 features from 27 sites (Table 1 and Fig. 7), was dominated by
Sarcobatus vermiculatus (greasewood) (Fig. 8) recovered at 20 sites.
It represents 28% of all identied charcoal. Pinus edulis (twoneedle
pinyon) is the second most abundant species recovered at 18 sites
throughout the canyon. This category also includes fragments only
identied as pine (Pinus), which most likely represent twoneedle
pinyon, a dominant species of this region. Although insignicant in
the pollen record, Juniperus (juniper) represents the third most
ubiquitous charcoal type identied in samples from 17 sites in Nine
Mile Canyon. Another conifer, Pseudotsuga menziesii (Douglas-r),
dominates the charcoal record from sites in the upper portion of
the canyon. Despite its absence or limited presence in assemblages
from the middle and lower portion of the Nine Mile Canyon study
area, Pseudotsuga menziesii (Douglas-r) was recovered at 15 sites
and totals 13.2 percent of all identied charcoal. Artemisia (sage-
brush) charcoal is the second most abundant (11.4%) hardwood
species, recovered at 15 sites, while Populus/Salix (cottonwood/
willow) charcoal fragments also were recovered in considerable
quantities (7.9%) at 7 sites. Eight sites yielded Amelanchier
(serviceberry) charcoal (2.3%) indicating the most common mem-
ber of rose family in this study. Other hardwood species (Fig. 9),
including Chrysothamnus (rabbitbrush), Atriplex (saltbush), Aster-
aceae (sunower family), Ephedra (jointr), Acer (maple), Rhus
(sumac), Rosaceae (rose family), and Cercocarpus (Mountain ma-
hogany) did not exceed 1% of the assemblage.
Fig. 6. Pollen diagram of percentage abundance for arboreal and shrubby woody taxa (excluding grasses, forbs, and culturally important taxa) identied in Nine Mile Canyon sites,
Utah (compiled by P. Kov
cik and Linda Scott Cummings).
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326318
5. Discussion
Various trees and shrubs with different ecological preferences,
noted in the pollen and charcoal records, reect the landscape
diversity of Nine Mile Canyon. Identied taxa were grouped into
four distinct groups (Table 1), based on the ecological divisions
known for the region (Albee et al., 1988; Bureau of Land
Management, 2011; Knight et al., 2010:108; Lanner, 1975, 1981,
1983; Mozingo, 1987; West, 1979) and observed in situ by the
authors. The major plant communities include montane conifer
forest, pinyon-juniper woodland, oodplain shrubland, and the
riparian zone. These ecological divisions were tested using Prin-
cipal Components Analysis (PCA) on charcoal percentages (Fig. 10).
Pseudotsuga menziesii (montane conifer forest) and taxa typical of
pinyon-juniper woodlands (Pinus/Pinus edulis,Juniperus, Rosaceae,
Cercocarpus,Ephedra, and Rhus) cluster on the left-hand side of
axis PC1. The right side of the same axis contains taxa typical of
oodplain shrubland (Sarcobatus vermiculatus, Asteraceae, Chrys-
othamnus, and Atriplex) as well as riparian species, including Sal-
icaceae (Populus/Salix) and Acer. Although Amelanchier is situated
on the right side but very close to the axis, its habitat is more
typical of dry slopes rather than of oodplains. In addition, a
strong correlation is observed between Juniperus and Pinus/Pinus
edulis, while Asteraceae species appear to form another group.
We aim to test the effects of the topographical characteristics
including elevation range and wall steepness of the six zones on
vegetation patterns observed in the pollen record and selection of
fuel/construction material during the Formative period in Nine
Mile Canyon.
5.1. Montane conifer forest
Fir, spruce, and Douglas-r, identied in the pollen record,
indicate montane conifer forests of the Tavaputs Plateau uplands
(~1800e3000 m asl). Presence of these taxa in the pollen record is
very sparse and indicates pollen blown in from higher elevations.
Of these, only Douglas-r is represented in the charcoal record,
especially in Zones I and II (Tables 2 and 3). Douglas-rgrowsin
elevations up to 2730 m, but it can be found on cooler north-facing
slopes in elevations as low as 1510 m. Samples from sites 42Cb3104
Fig. 7. Anthracological diagram of percentage fragment counts of key taxa identied in Nine Mile Canyon sites, Utah (compiled by P. Kov
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326 319
Fig. 8. SEM microphotographs of wood charcoal specimens from archaeological sites in Nine Mile Canyon: (A) Sarcobatus vermiculatus (greasewood) transverse section (TS), (B)
Pinus edulis (twoneedle pinyon) transverse section (TS), (C) Juniperus (juniper) transverse section (TS), (D) Pseudotsuga menziesii (Douglas-r) transverse section (TS), (E) Pseu-
dotsuga menziesii (Douglas-r) tangential section (TLS), (F) Artemisia (sagebrush) transverse section (TS), (G) Populus (cottonwood/aspen) transverse section (TS), (H) Salix (willow)
transverse section (TS), and (I) Amelanchier (serviceberry) transverse section (TS) (photo by P. Kov
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326320
and 42Cb3105 (Zone II) contained more than 50% of Douglas-r
charcoal, indicating easy access and availability of this taxa.
Absence of Douglas-r charcoal in Zone V (Table 4) and its limited
presence in the assemblages from Zones IV (Table 5) and VI
(Table 6) is expected due to topography of the canyon at these lo-
cations, consisting of tall cliffs (Fig. 11). Elevated Douglas-r char-
coal values at site 42Dc3053 (Zone IV) may indicate burned
construction wood. In contrast to the majority of single hearth sites
in the study area, 42Dc3053 represent one of the largest sites with
several superimposed pit structures. Douglas-r provides one of
the most valuable timbers for construction in this area. It is possible
that in areas with limited access to this wood type, use of Douglas-
r as fuel might be restricted to waste pieces such as structural
elements that needed to be replaced or were discarded due to wear,
damage, or infestation by insects/termites. Burning drift wood or
wood collected from debris-ow deposits on the canyon oor also
could have introduced this type of wood in Zone IVeVI charcoal
5.2. Pinyon-juniper woodland
South and east-facing slopes of the canyon support pinyon-
juniper woodlands and shrubland communities, represented by
twoneedle pinyon and juniper. Both pollen and charcoal records
indicate pinyon/pine as one of the most available and frequently
used conifers in the canyon. The palynological record indicates a
decreasing trend of pine pollen concentration downstream along
Fig. 9. SEM microphotopraphs of wood charcoal specimens from archaeological sites in Nine Mile Canyon: (A) Chrysothamnus (rabbitbrush) transverse section (TS), (B) Atriplex
(saltbush) transverse section (TS), (C) Ephedra (jointr) transverse section (TS), (D) Acer (maple) transverse section (TS), (E) Rhus (sumac) transverse section (TS), and (F) Cercocarpus
(mountain mahogany) transverse section (TS) (photo by P. Kov
Fig. 10. Principal Components Analysis (PCA) plot of anthracological dataset from sites
in Nine Mile Canyon, Utah.
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326 321
the Nine Mile Creek. Variability in abundance of twoneedle pinyon
charcoal at sites from different locations in the canyon may reect
accessibility to this tree. Absence or limited presence of pinyon
charcoal fragments in the assemblages of three neighboring sites
(42Cb0683, 42Cb3036, and 42Cb3035) in Zone V demonstrates
how the local topography, represented by steep cliffs rising above
the canyon bottom, affects species representation in the charcoal
assemblages. In addition, pine bark scale (a byproduct of burning
pine logs), commonly recovered at the sites from this study area,
was absent in the macrooral records from these three sites, also
suggesting limited pinyon growth in this area.
Pinyon trees produce edible nuts collected by prehistoric oc-
cupants of the canyon. Charred pine cone scales and seed fragments
were recovered from site 42Cb3075 (Zone I) while charred pine
seeds were noted in six features/samples from site 42Dc3053 (Zone
IV), indicating processing pine nuts. Recovery of pine cone scales
suggests local nut processing and, hence, probably local growth of
Juniper charcoal fragments also were present in anthracological
assemblages from all zones of the canyon. Juniper charcoal gener-
ally accompanies the appearance of pinyon charcoal, except for site
42Cb3036, where the charcoal record is exclusively composed of
juniper. Unlike charcoal, the pollen record provides limited juniper
pollen concentration, probably resulting from production of lesser
quantities of pollen per tree, as well as a greater tendency towards
juniper pollen deterioration with time.
Other species typical of this environment include jointr,
members of the rose family (serviceberry and mountain mahog-
any), sumac, and oak. Oak, noted only in the pollen record, appears
to be insignicant or inaccessible for prehistoric populations.
5.3. Floodplain shrubland
Although greasewood and sagebrush both represent oodplain
desert shrub communities, various sagebrush species grow in ele-
vations ranging between 820 and 3400 m (Albee et al.,
1988:37e41). On the other hand, greasewood prefers alkaline or
saline soils (Mozingo,1987) and grows at elevations of 840e2120 m
(Albee et al., 1988:219). These ecological preferences were observed
in the pollen record, showing decreasing sagebrush and increasing
greasewood pollen downstream of the creek. The charcoal record
conforms to this trend. Greasewood charcoal is completely absent
at sites from Zone I (~2080 m asl), located along the upper portion
of the Nine Mile Creek, while it is continuously recovered at sites
from Zones III (Table 7) through VI. Greasewood was the only type
identied in the charcoal assemblage from site 42Cb3117 (Zone II),
while at eight other sites it comprises more than 56% of the record.
Abundance of this taxon indicate that most sites were located
above the active oodplain, where soil is more saline, leaving rst
terraces available for corn agriculture. Other salt-tolerant taxa
include saltbush and other members of the goosefoot and
amaranth family. Saltbush charcoal was recovered only at sites
from Zone VI, while charred saltbush seeds and fruits were noted in
all zones, except Zone I. In addition to saltbush, charred seeds
recovered from different sites in the canyon indicate amaranth and
goosefoot processing. The pollen record also reects a signicant
increase of Amaranthaceae pollen grains downstream in the
Although sagebrush charcoal does not follow the pattern
observed in the pollen record, its high abundance (above 75%) at
three sites from Zones II, IV, and VI is interesting. Two of these
Table 4
Fragment count (C), percentage abundance (%), and ubiquity (U) for anthracological assemblages from Zone V (compiled by P. Kov
Zone V
~1720-1710 m asl
Plant community
Taxon Site Totals
42Cb0683 42Cb3036 42Cb3035
Scientic Name Common Name C % U C % U C % U C % U
Pinyon-juniper Juniperus Juniper 40 100.0 100.0 40 25.0 25.0
Pinus/Pinus edulis Pine/Twoneedle pinyon 1 1.3 50.0 1 0.6 25.0
Amelanchier Serviceberry 12 30.0 100.0 12 7.5 25.0
Floodplain Artemisia Sagebrush 2 2.5 50.0 2 1.3 25.0
Sarcobatus vermiculatus Greasewood 28 70.0 100.0 76 95.0 100.0 104 65.0 75.0
Riparian Acer Maple 1 1.3 50.0 1 0.6 25.0
Totals 40 100.0 40 100.0 80 100.0 160 100.0
# of Taxa 2 1 4 6
# of Features/Samples 1 1 2 4
Table 5
Fragment count (C), percentage abundance (%), and ubiquity (U) for anthracological assemblages from Zone IV (compiled by P. Kov
Zone IV
~1790e1830 m asl
Plant community
Taxon Site Totals
42Dc3053 42Dc3147 42Dc3145
Scientic Name Common Name C % U C % U C % U C % U
Montane Pseudotsuga menziesii Douglas-r 151 18.0 81.8 1 1.3 50.0 152 16.0 76.0
Pinyon-juniper Juniperus Juniper 93 11.1 59.1 24 30.0 100.0 117 12.3 60.0
Pinus/Pinus edulis Pine/Twoneedle pinyon 151 18.0 86.4 41 51.3 100.0 192 20.2 84.0
Amelanchier Serviceberry 9 1.1 22.7 9 0.9 20.0
Floodplain Asteraceae Sunower family 9 1.1 18.2 9 0.9 16.0
Artemisia Sagebrush 92 10.9 77.3 25 83.3 100.0 117 12.3 72.0
Chrysothamnus Rabbitbrush 9 1.1 18.2 1 1.3 50.0 10 1.1 20.0
Sarcobatus vermiculatus Greasewood 201 23.9 77.3 5 16.7 100.0 9 11.3 50.0 215 22.6 76.0
Riparian Populus/Salix Cottonwood/Willow 126 15.0 81.8 4 5.0 50.0 130 13.7 76.0
Totals 841 100.0 30 100.0 80 100.0 951 100.0
# of Taxa 9 2 6 9
# of Features/Samples 22 1 2 25
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326 323
features represent roasting pits (sites 42Cb3112 and 42Dc3147),
which may explain selection of a certain type of wood, such as
sagebrush, to add a avor. Sagebrush also was used ceremonially
(Dunmire and Tierney, 1997:193; Moerman, 1998:92e103). Other
members of the Asteraceae family, including rabbitbrush, have
similar environmental preferences as sagebrush, and often grow in
the same habitat.
5.4. Riparian zone
Trees and shrubs of the riparian plant community are indicated
only in the anthracological record by cottonwood/willow and a few
maple charcoal fragments. The cottonwood/willow fragments
might represent aspen, which grows mixed with Douglas-rat
higher elevations. The likelihood that Populus/Salix charcoal rep-
resents aspen is greater at site 42Cb3062 than at sites downstream,
as assessed by recovery of a greater quantity of Pseudotsuga
(Douglas-r) charcoal there and this site's placement in Zone II.
Charred willow twigs identied at site 42Dc3053 indicate willow
was used in construction of a mat exposed on the oor of a pit
structure. In addition, cattail pollen and charred seeds, as well as
pollen grains reecting a member of the sedge family, also indicate
riparian vegetation.
6. Conclusions
Anthracological and pollen analysis conducted on feature ll
recovered from multiple sites in Nine Mile Canyon, Utah, western
United States, provide a new view for reconstruction of prehistoric
woodlands. Four major vegetation communities, montane forest,
pinyon-juniper woodland, oodplain shrubland, and riparian zone,
were identied in the vicinity of sites situated along the upper and
middle portion of the Nine Mile Creek.
This study indicates that distribution of different plant com-
munities correlates to elevation and canyon wall steepness
(Fig. 11). Conifers, including Douglas-r, twoneedle pinyon, and
juniper, dominated both pollen and charcoal assemblages from
sites in the upper portion of the canyon with moderately steep
slopes. The pollen record indicates conifer (especially pine)
regression at sites in the downstream portion of the canyon, while
oodplain species including greasewood and members of the
amaranth/goosefoot family increase in abundance. These obser-
vations also are reected in the anthracological record, suggesting
prehistoric people used plants available to them directly in the
vicinity of their settlements. However, recovery of Douglas-r
charcoal from sites in the lower portion of the canyon's study
area indicates possible use of driftwood or wood collected from
the debris-ows for fuel. Distribution of sagebrush and grease-
wood observed in pollen and charcoal records indicates higher
salinity of the oodplain in the lower portion of the canyon.
Overall, lower quantities of sagebrush in the pollen record and in
anthracological assemblages from the lower reaches of the Nine
Mile Creek study area, compared with quantities observed on the
landscape today, may reect clearing activities and use of the
active oodplain or alluvial fans for agriculture.
In some cases these results indicated selection of certain wood
types by site occupants. Charcoal assemblages from hearth fea-
tures associated with pit structures at site 42Dc3053 suggested
secondary use of structural elements made of Douglas-r timber
as a fuel. Willow twigs were used in construction of a mat
encountered at the same site. Perhaps sagebrush wood was used
to add avor to food prepared in roasting pits from sites 42Cb3112
and 42Dc3147, although it simply might have been readily avail-
able locally. These examples suggest human preference based on
wood properties, including construction purposes as well as food
Table 6
Fragment count (C), percentage abundance (%), and ubiquity (U) for anthracological assemblages from Zone VI (compiled by P. Kov
Zone VI
~1630 m asl
Plant community
Taxon Site Totals
42Cb2909 42Cb3069 42Cb3070 42Cb3071 42Cb3072 42Cb2910 42Cb3074
Scientic Name Common Name C % U C % U C % U C % U C % U C % U C % U C % U
Montane Pseudotsuga menziesii Douglas-r 2 5.4 100.0 1 2.5 100.0 10 2.9 37.5 13 2.0 29.4
Pinyon-juniper Juniperus Juniper 8 9.9 100.0 1 1.8 50.0 2 5.4 100.0 6 15.0 100.0 17 41.5 100.0 50 14.6 87.5 84 13.2 82.4
Pinus/Pinus edulis Pine/Twoneedle pinyon 30 37.0 66.7 5 13.5 100.0 20 50.0 100.0 13 31.7 100.0 67 19.6 87.5 135 21.2 70.6
Ephedra Jointr 5 6.2 33.3 5 0.8 5.9
Rosaceae Rose family 1 1.2 33.3 1 0.3 12.5 2 0.3 11.8
Amelanchier Serviceberry 7 8.6 66.7 11 3.2 37.5 18 2.8 29.4
Floodplain Asteraceae Sunower family 1 1.8 50.0 1 0.2 5.9
Artemisia Sagebrush 10 12.3 66.7 2 3.6 100.0 1 2.7 100.0 4 9.8 100.0 61 17.8 100.0 30 75.0 100.0 108 17.0 88.2
Chrysothamnus Rabbitbrush 9 22.5 100.0 9 1.4 5.9
Sarcobatus vermiculatus Greasewood 20 24.7 66.7 51 92.7 100.0 21 56.8 100.0 12 30.0 100.0 6 14.6 100.0 109 31.9 100.0 219 34.4 88.2
Atriplex Saltbush 6 16.2 100.0 1 2.5 100.0 1 2.4 100.0 5 1.5 25.0 1 2.5 100.0 14 2.2 35.3
Riparian Acer Maple 1 0.3 12.5 1 0.2 5.9
Populus/Salix Cottonwood/Willow 27 7.9 62.5 27 4.2 29.4
Totals 81 100.0 55 100.0 37 100.0 40 100.0 41 100.0 342 100.0 40 100.0 636 100.0
# of Taxa 7465510313
# of Features/Samples 3 2 1 1 1 8 1 17
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326324
Combined pollen and anthracological analyses, as presented
here, has the potential to investigate differences in vegetation that
might accompany long-term occupation and agricultural activities
and/or climate change through time when applied to samples
collected at close stratigraphic intervals. We recommend these
analyses be paired in future archaeological studies throughout the
American Southwest and adjoining regions.
Funding for this research was provided by Carbon County,
Duchesne County, and Jones and DeMille Engineering. We thank
Jody Patterson (Montgomery Archaeological Consultants) for
providing feature soils for analysis, editorial input, and acting as a
guide through Nine Mile Canyon. We recognize contributions to
this dataset by Kathryn Puseman and R.A. Varney as well as
constructive comments and suggestions from Jennifer Milligan.
Additionally, we thank the two anonymous reviewers for providing
suggestions and comments that help improve the article.
Albee, B.J., Shultz, L.M., Goodrich, S., 1988. Atlas of the Vascular Plants of Utah. in:
Utah Museum of Natural History Occasional Publication No.7. The Utah
Fig. 11. Archaeological sites (black dots) from a section of Nine Mile Canyon cluster in six different zones (IeVI). The longitudinal prole (blue line) shows the height (meters) of the
creek bed above sea level as well as the gradient and length (kilometers) of Nine Mile Creek in the study area. Approximate transverse proles of Nine Mile Canyon for each zone
show increasing wall steepness from the upper to lower reach of the creek. Pie charts display proportions of different plant communities indicated in the pollen andanthracologocal
records. A gradual increase of oodplain scrubland habitats and decrease of pinyon-juniper woodlands toward the lower reaches of the canyon is evident. In the pollen record, the
apparent elevated proportion of oodplain shrubland category in Zone I likely indicates inclusion of pollen from upland sagebrush (Artemisia spp.). Artemisia pollen is not iden-
tiable to species, so the upland and oodplain shrubland taxa cannot be separated. Douglas-r wood (montane habita t) appears to be used mostly in the upper (Zones I and II) and
middle (Zones III and IV) reaches of the Nine Mile Creek study area. The absence or limited presence of this taxon in the lower reaches of the canyon's study area correlates with
canyon wall steepness in Zones V and VI (see transverse proles) (compiled by P. Kov
cik). (For interpretation of the references to colour in this gure legend, the reader is referred
to the web version of this article.)
Table 7
Fragment count (C), percentage abundance (%), and ubiquity (U) for anthracological assemblages from Zone III (compiled by P. Kov
Zone III
e1830 m asl
Taxon Site Totals
42Dc3364 42Dc3365 42Cb3117 42Dc3366 42Dc3367 42Dc3368
Scientic Name Common Name C % U C % U C % U C % U C % U C % U C % U
Montane Pseudotsuga
Douglas-r 2 4.9 50.0 2 7.7 100.0 22 15.2 75.0 2 5.0 100.0 28 8.4 60.0
Juniperus Juniper 3 7.3 100.0 8 30.8 100.0 16 11.0 100.0 27 8.1 70.0
Pinus/Pinus edulis Pine/Twoneedle
36 87.8 50.0 5 19.2 100.0 4 10.0 100.0 48 33.1 100.0 3 7.5 100.0 96 28.9 80.0
Amelanchier Serviceberry 7 4.8 25.0 7 2.1 10.0
Floodplain Artemisia Sagebrush 8 5.5 50.0 8 2.4 20.0
Chrysothamnus Rabbitbrush 3 2.1 25.0 1 2.5 100.0 4 1.2 20.0
Greasewood 11 42.3 100.0 40 100.0 100.0 32 80.0 100.0 18 12.4 100.0 27 67.5 100.0 128 38.6 80.0
Riparian Populus/Salix Cottonwood/
4 10.0 100.0 23 15.9 75.0 7 17.5 100.0 34 10.2 50.0
Totals 41 100.0 26 100.0 40 100.0 40 100.0 145 100.0 40 100.0 332 100.0
# of Taxa 3 4 1 3 8 5 8
# of Features/
2111 41 10
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326 325
Museum of Natural History. Salt Lake City.
Asouti, E., 2003. Woodland vegetation and fuel exploitation at the prehistoric
campsite of Pinarbas¸ i, south-central Anatolia, Turkey: the evidence from the
wood charcoal macro-remains. J. Archaeol. Sci. 30, 1185e1201.
Asouti, E., 2013. Woodland vegetation, rewood management and woodcrafts at
Neolithic Çatalh
oyük. In: Hodder, I. (Ed.), Humans and Landscapes of Çata-
oyük. Cotsen Institute of Archaeology Press, Los Angeles, CA, pp. 129e161.
Asouti, E., Austin, P., 2005. Reconstructing woodland vegetation and its exploitation
by past societies, based on the analysis and interpretation of archaeological
wood charcoal and macro-remains. Environ. Archaeol. 10, 1e18.
Asouti, E., Kabukcu, C., White, C.E., Kujit, I., Finlayson, B., Makarewicz, C., 2015. Early
Holocene woodland vegetation and human impacts in the arid zone of the
southern Levant. Holocene 25 (10), 1565e1580.
Baillie, M.G.L., 1982. Tree-ring Dating and Archaeology. The University of Chicago
Press, Chicago.
Blinnikov, M.S., 2005. Phytoliths in plants and soils of the interior Pacic Northwest.
USA. Rev. Palaeobot. Palynology 135, 71e98.
Bureau of Land Management, 2011. West Tavaputs Plateau natural gas full eld
development plan; nal environmental impact statement UT-070-05-055. Bu-
reau of Land Management, Price Field Ofce.
Carlquist, S., 2001. Comparative Wood Anatomy: Systematic, Ecological, and
Evolutionary Aspects of Dicotyledon Wood, second ed. Springer Series in Wood
Science. Springer, Berlin.
Dunmire, W.W., Tierney, G.D., 1997. Wild Plants and Native Peoples of the Four
Corners. Museum of New Mexico Press, Sante Fe.
Dussol, L., Elliott, M., Pereira, G., Michelet, D., 2016. The use of rewood in ancient
Maya funerary rituals: a case study from Rio Bec (Campeche, Mexico). Lat. Am.
Antiq. 27 (1), 51e73.
Elliott, M., 2012. An anthracological approach to understanding Late Classic period
cultural collapse in Mesoamerica's northwestern frontier. In: Badal, E.,
Carrion, Y., Macias, M., Ntinou, M. (Eds.), Wood and Charcoal: Evidence for
Human and Natural History. Saguntum Extra, vol. 13, pp. 217e225.
Emery-Barbier, A., Thi
ebault, S., 2005. Preliminary conclusions on the Late Glacial
vegetation in south-west Anatolia (Turkey): the complementary nature of
palynological and anthracological approaches. J. Archaeol. Sci. 32 (8),
Gillin, J., 1938. Archaeological Investigations in Nine Mile Canyon, Utah (During the
Year 1936). Bulletin of the University of Utah 28(11).
Gillin, J., 1955. Archeological Investigations in Nine Mile Canyon, Utah: a Re-pub-
lication. University of Utah Anthropological Papers 21, pp. 1e32.
Henry, A., Th
ery-Parisot, I., 2014. Fuel use and management during the Mesolithic:
recent approaches in archaeobotany. Toulouse (France), Palethnology. In:
Henry, A., Marquebielle, B., Chesnaux, L., Michel, S. (Eds.), Techniques and
Territories: New Insights into Mesolithic Cultures. Proceedings of the Round
Table, November 22-23, 2012, vol. 6, pp. 65e83.
Hoadley, B., 1990. Identifying Wood: Accurate Results with Simple Tools. The
Taunton Press, Inc., Newtown.
Klimaszewski-Patterson, A., Mensing, S.A., 2016. Multi-disciplinary approach to
identifying Native American impacts on Late Holocene forest dynamics in the
southern Sierra Nevada range, California, USA. Anthropocene 15, 37e48.
Knight, T.A., Meko, D.M., Baisan, C.H., 2010. A bimillennial-length tree-ring recon-
struction of precipitation for the Tavaputs Plateau, Northeastern Utah. Quat.
Res. 73 (1), 107e117.
cik, P., Scott Cummings, L., Puseman, K., Dalpra, C., Milligan, J.L.B., 2014. Pollen,
Macrooral, Protein Residue, and Organic Residue (FTIR) Analysis of Samples
from Sites along Nine Mile Canyon. Utah. PRI Technical Report 2013-065. Ms. on
le with Montgomery Archaeological Consultants, Moab, UT.
Lanner, R.M., 1975. Pinon pines and junipers of the southwestern woodlands. In:
The Pinyon-juniper Ecosystem: a Symposium. Utah State University, College of
Natural Resources Utah Agricultural Experiment Station, Logan, UT, pp. 1e18.
Lanner, R.M., 1981. The Pi~
non Pine: a Natural and Cultural History. University of
Nevada Press, Reno.
Lanner, R.M., 1983. Trees of the Great Basin. University of Nevada Press, Reno.
Louderback, L.A., Rhode, D.E., 2009. 15,000 years of vegetation change in the Bon-
neville Basin: the Blue Lake pollen record. Quat. Sci. Rev. 28, 308e326.
Louderback, L.A., Rhode, D., Madsen, D.B., Metcalf, M., 2015. Rapid vegetation shifts
in the uinta mountains (Utah and Wyoming, USA) during the late Pleistocene
and holocene. Palaeogeography, Palaeoclimatology, Palaeoecology 438,
Marston, J.M., 2009. Modeling wood acquisition strategies from archaeological
charcoal remains. J. Archaeol. Sci. 36, 2192e2200.
Marwitt, J.P., 1986. Fremont cultures. Sturtevant, W. C. (Gen. Ed.), Handbook of
North American Indians. In: D'Azevado, W.L. (Ed.), Great Basin, vol. 11. Smith-
sonian Institution, Washington, D.C, pp. 161e172.
Matthews, M.H., 1979. Soil sample analysis of 5MT2148: dominguez ruin, dolores,
Colorado. Appendix B. In: Reed, A.D. (Ed.), The Dominguez Ruin: a McElmo
Phase Pueblo in Southwestern Colorado. Cultural Resource Series No. 7. Bureau
of Land Management, Denver.
Minnis, P.E., 1978. Early Prehistoric Ethnobotany in Chaco Wash: Plant Remains
from the Tsaya Project, New Mexico. Ms. On File with Laboratory of Anthro-
pology Museum of New Mexico (Santa Fe, New Mexico).
Minnis, P.E., 1987. Identication of wood from archaeological sites in the American
Southwest. I. Keys for gymnosperms. J. Archaeol. Sci. 14, 121e131.
Minnis, P.E., Raymer, L.E., 1982. Macroplant Remains from Curecanti National Rec-
reation Area, Colorado. Ms. On File with Midwest Archaeological Center. Na-
tional Park Service.
Moerman, D.E., 1998. Native American Ethnobotany. Timber Press, Inc., Portland.
Morss, N., 2009. The Ancient Culture of the Fremont River in Utah: Report on the
Explorations under the Clain-Emerson Fund, 1928-1929. The University of
Utah Press, Salt Lake City.
Mozingo, H.N., 1987. Shrubs of the Great Basin. University of Nevada Press, Reno.
adi, K., Sümegi, P., T
ocsik, T., 2012. Charcoal and Pollen analyses and vege-
tation reconstruction of the Alpine foreland in West Hungary. Central Eur. J.
Geosciences 4 (4), 592e602.
Nelle, O., Dreibrodt, S., Dannath, Y., 2010. Combining pollen and charcoal: evalu-
ating Holocene vegetation composition and dynamics. J. Archaeol. Sci. 37,
Nocus, N., Wiethold, J., Ertlen, D., Schneider, N., Richard, H., 2011. First anthraco-
logical results from Rhine's Plain and comparison with other palaeo-
environmental data. Sagvntvm Extra 11, 141e142.
Novak, J., Jankovsk
a, V., Be
sta, T., Dreslerov
a, D., 2011. Long term vegetation changes
in the Bilina River region, Czech Republic. In: Badal, E., Carrion, Y., Grau, E.,
Macias, M., Ntinou, M. (Eds.), The Charcoal as Cultural and Biological Heritage.
Saguntum Extra, vol. 11, pp. 139e140.
Panshin, A.J., de Zeeuw, C., 1980. Textbook of Wood technology. McGraw-Hill Book,
Co., New York.
Patterson, J.J., 2015. Maize growing, processing, and storage: evidence from nine
mile canyon, Arizona. Archaeol. Southwest Mag. 29, 24e25.
Patterson, J.J., 2016. Formative period changes in regional interaction and inuence
in Nine Mile Canyon, Utah. In: Paper Presented at the 81st Annual Meeting of
the Society for American Archaeology. Florida, USA, Orlando.
Patterson, J.J., Flanigan, T.H., 2010. Capacity, distribution, and spatial associations of
granaries in Nine Mile Canyon, a historical ecology perspective on arable
acreage and mobility. In: Utah. Paper Presented at the 32nd Great Basin
Anthropological Conference. Layton, Utah, USA.
Picornell-Gelabert, L., Asouti, E., Allue
e, E., 2011. The ethnoarchaeology of rewood
management in the Fang villages of Equatorial Guinea, central Africa: impli-
cations for the interpretation of wood fuel remains from archaeological sites.
J. Anthropol. Archaeol. 30, 375e384.
Puseman, K., Scott Cummings, L., 2013. Pollen and Macrooral Analysis of Samples
from Sites along Nine Mile Canyon, Carbon County, Utah. PRI Technical Report
2013-040. Ms. on le with Montgomery Archaeological Consultants, Moab,
Salavert, A., Dufraisse, A., 2014. Understanding the impact of socio-economic ac-
tivities on archaeological charcoal assemblages in temperate areas: a compar-
ative analysis of rewood management in two Neolithic societies in western
Europe (Belgium, France). J. Anthropol. Archaeol. 35, 153e163.
Scheel-Ybert, R., 2001. Man and vegetation in southeastern Brazil during the late
holocene. J. Archaeol. Sci. 28, 471e480.
Schweingruber, F.H., Borner, A., Schulze, E., 2011. Atlas of Stem Anatomy in Herbs,
Shrubs and Trees Vol I. Springer-Verlag, Berlin Heidelberg.
Schweingruber, F.H., Borner, A., Schulze, E., 2013. Atlas of Stem Anatomy in Herbs,
Shrubs and Trees Vol. II. Springer-Verlag, Berlin Heidelberg.
Shackleton, C.M., Prins, F., 1992. Charcoal analysis and the principle of least effort e
a conceptual model. J. Archaeol. Sci. 19, 631e637.
Spangler, J.D., 2000a. One-pot pithouses and Fremont paradoxes: formative stage
adaptations in the Tavaputs plateau region of northeastern Utah. In:
Madsen, D.B., Metcalf, M.D. (Eds.), Intermountain Archaeology. The University
of Utah Press, Salt Lake City, UT, pp. 25e38.
Spangler, J.D., 2000b. Radiocarbon dates, acquired wisdom, and the search for
temporal order in the Uinta Basin. In: Madsen, D.B., Metcalf, M.D. (Eds.),
Intermountain Archaeology. The University of Utah Press, Salt Lake City, UT,
pp. 48e68.
Spangler, J.D., 2013. Nine Mile Canyon; the Archaeological History of an American
Treasure. The University of Utah Press, Salt Lake City.
Spangler, J.D., Spangler, D.K., 2010. Treasures of the Tavaputs: the Archaeology of
Desolation Canyon, Nine Mile Canyon and Range Creek. Colorado Plateau
Archaeological Alliance, Ogden.
West, N.E., 1979. Basic synecological relationships of sagebrush-dominated lands in
the Great Basin and the Colorado Plateau. In: The Sagebrush Ecosystem: A
Symposium. Utah State University, Logan, pp. 33e41.
Whitlock, C., Larsen, C., 2001. Charcoal as a re proxy. In: Smol, J.P., Birks, H.J.B.,
Last, W.M. (Eds.), Tracking Environmental Change Using Lake Sediments, Vol. 3:
Terrestrial, Algal, and Siliceous Indicators. Kluwer Academic Publishers, Dor-
drecht, Netherlands.
Wright, N.J., Fairbairn, A.S., Faith, T.J., Matsumura, K., 2015. Woodland modication
in bronze and iron age central anatolia: an anthracological signature for the
hittite state? J. Archaeol. Sci. 55, 219e230.
Wright, N.J., Fairbairn, A.S., Üstünkaya, M.C., Faith, J.T., 2017. Explaining changing
patterns of wood presence across the Bronze and Iron Age at Kaman-
oyük, central Anatolia. Quat. Int. 431, 90e102.
P. Kov
cik, L. Scott Cummings / Quaternary International 463 (2018) 312e326326
... A total of 204 flotation samples were examined for charcoal, with 61 containing fragments larger than 2 mm, alongside an additional 26 hand-collected samples. All samples were identified completely up to a maximum of 40 fragments (after Kováčik and Cummings 2018); only seven samples contained more fragments, and every one of these seven samples contained only oak charcoal. Hand-picked charcoal was examined in the same manner as flotation samples, but the two datasets are presented independently, as we assume that hand-picked samples likely represent only one or a few original charcoal pieces that were further fragmented during collection, transport, and analysis. ...
Full-text available
The Middle and Late Bronze Ages of western Anatolia (modern Turkey) remains poorly understood in comparison with its Mycenaean and Hittite neighbours, especially in agricultural economies and land use. Kaymakçı is the largest Middle and Late Bronze Age citadel excavated to date in western Anatolia and new archaeobotanical and zooarchaeological data from the site presented here shed new light on regional agricultural economies and land use. Agricultural practices at Kaymakçı focused on barley and bitter vetch farming and pig, caprine, and cattle husbandry within a diverse and extensive economic system that made substantial use of wild plants and animals for food, technology, and fuel. Goats and pigs were managed primarily for meat, while sheep and cattle were managed to produce a range of secondary products. Wood charcoal analysis reconstructs both deciduous and evergreen oak woodlands, which also dominate the contemporary landscape. In regional perspective, Kaymakçı is most similar to the northern Aegean agricultural tradition, but with elements of Anatolian practices as well, representing a hybrid position between the Aegean and Anatolian worlds as seen in other lines of archaeological evidence from the site.
... Wood has long played a significant role in human livelihoods and social organization as a raw material for construction, cooking, fire making, heating, and artifact production (Rubiales et al., 2011;Salavert and Dufraisse, 2014;Wang et al., 2014a,b;Medina-Alcaide et al., 2015;Kabukcu, 2018). Charcoal remains derived from cultural features with clear functional attributes are the result of conscious choice and can reflect information on wood use and the patterns of its collection in the past Marcos and Ortega, 2014;Wang et al., 2014a;Rhode, 2016;Hazell et al., 2017;Kováčik and Cummings, 2017;Mafferra, 2017;Mota and Scheel-Ybert, 2019;Kabukcu and Chabal, 2020). Tamarix L. was the most abundant taxon in houses, hearths, and ash pits at JBR. ...
Full-text available
Charcoal remains from archeological sites are used not only to reconstruct the historical composition of local woodlands but also to examine the history of the human use of wood. Nevertheless, key questions such as how and why people may have selected particular woody taxa from locations long distances from their habitat have rarely been addressed. In the present study, we analyze charcoal remains from the ancient Jingbaoer (JBR) jade mine in the Mazong Mountains (Mazong Shan) of Northwest China to explore patterns in the collection and use of wood by Iron Age people. Factors affecting the choice of wood collected at the JBR site are discussed by combining the results of pollen records and charcoal analysis. Our results suggest that tamarisk (Tamarix L.), a shrub dominant in the local area, was the main source of wood for JBR miners and was used as firewood depending upon its local availability. The miners may also have used wood from species sourced further away, such as Pinus L. and Picea L., because of the local scarcity of these trees in such a dry environment. The agropastoralist subsistence system practiced by the JBR miners supports the hypothesis of the collection of wood from distant locales. This study highlights diverse patterns of wood collection in an area scarce in woody plants and provides new evidence for understanding how Iron Age people adapted to extremely arid environments.
... Initially we planned to identify at least 40 charcoal fragments from each sample after Kováčik and Cummings (2018), however, not all samples contained 40 charcoal fragments. Only 46 of the 104 samples (44%) contained 40 or more charcoal fragments. ...
Çamlıbel Tarlası is a short-lived, mid 4th millennium BCE Chalcolithic archaeological site in northern central Anatolia, modern Turkey, with evidence for both intensive metallurgy and permanent occupation. Analysis of a wood charcoal assemblage from the site, totaling 2815 charcoal fragments, is the first from this period and region. Anthracological analysis indicates that the primary fuel wood used was deciduous oak, which comprised nearly 90% of identifiable fragments. We find little evidence of differences in wood species used for different functions or over time; however, a significant trend towards the increased use of large-diameter branch or trunk wood over time is noted both for oak and other minor taxa. We reconstruct a dense oak-dominated woodland in the vicinity of the site at the time of first use, with increased forest clearance over time, due to either diminished fuel availability or agricultural expansion, or a combination of the two. An intensification in metallurgical activity in later periods of occupation may have increased demand specifically for large-diameter wood.
... The integration of charcoal and pollen analysis is of partic- ular interest to vegetation reconstruction, counterbalancing their disadvantages and combining their advantages ( Nelle et al., 2010). Here in this issue, contributions combine at least two proxies: pol- len and wood charcoal in Nine Mile Canyon, Utah (Kov a cik and Scott Cummings, 2017), from Borsod plain in North Hungary (Moskal-del Hoyo et al., 2017a) and in NE Iberia (Piqu e et al., 2017); charcoal, wood, and pollen and plant remains from camelid and human coprolites from Patagonia ; archaeological and soil charcoals compared with pollen in the Litovelsk e Pomoraví region, Czech Republic (Nov ak et al., 2017); waterlogged wood, charcoal, strontium isotopes, and pollen in southwestern Germany ( Million et al., 2017); archaeology, charcoal, pollen, vegetation surveys and historic documents, in Cinque Terre National Park, Italy ( Pescini et al., 2017); wood charcoal and carpo- logical remains at Cova de les Cendres, Spain (Badal García and Martínez Varea, 2017); and Uzquiano (2017) compared her anthra- cological data with available pollen diagrams. ...
Full-text available
Human activities are usually considered to have had impacts on forest degradation and/or changes of forest composition in the past. The Tibetan Plateau, an environmentally sensitive area, has witnessed a high intensity of human activities since the late Neolithic, but the evidence is still incomplete concerning how human activities affected the environment, especially vegetation changes. This may result, in part, from the fact that research on prehistoric wood utilization by humans in the northeast Tibetan Plateau (NETP) is scarce, especially from the perspective of charcoal analysis, which is an archaeological index directly related to human activities. To fill this gap, we report new data here concerning charcoal analysis from 24 archaeological sites dated 5200-2300 a BP in the NETP, where this period saw remarkable environmental and social changes. By coupling the results with published paleoenvironment records (especially pollen), we aim to explore the utilization of wood and its spatial-temporal changes, as well as the anthropogenic impacts on past forest dynamics in the NETP. The results suggest that the wood resources collected by humans were mainly needleleaf trees (Picea L. and Pinus L.), broadleaved trees (Populus L., Salix L., and Betula) as well as shrubs (Hippophae L. and Tamarix L.), and that these were distributed around the human habitation areas. There was a clear spatial difference in the woody assemblage collected by humans in the period 3600-2300 a BP. This was likely related to the diversity of natural environments in local areas, but there was an abnormal spatial pattern of forest composition with broadleaved trees dominant above 2500 masl caused not by climate change, but by the rapid increase of human settlement and wood collection in high altitude areas of the NETP since 3600 a BP.
European colonization brought innumerable changes and choices to Native groups across the Southeast. Scholars continue to examine the various ways communities navigated these disruptions. Studying the remains of daily practice offers a window into how communities negotiated continuity and change. Wood charcoal, representing the remains of daily fires, provides an important, but underutilized, method for examining people’s daily routines and interactions with their surrounding landscapes. This paper examines wood charcoal assemblages from several sites in the North Carolina Piedmont that span the precontact to early colonial periods (AD 1400–1705). Fuelwood collection models are used to consider the environments, practices, and preferences that influenced the composition of wood charcoal assemblages. Comparison of these datasets shows a consistent significant pattern of high-quality fuelwood selection with additional patterns potentially related to long-term use of the same environment and factors related to colonialism. Altogether, these patterns suggest continuity of some daily practices despite disruptions to other aspects of life.
Full-text available
For over 50 years, researchers have suggested that increased regional rainfall over the highland deserts of Mesoamerica’s northwestern frontier zone during the Classic period (AD 200-900) allowed for the colonization of the zone by farming groups who originated from Central and/or West Mexico. A severe and prolonged drought is hypothesized to have later provoked the abandonment of the region by these sedentary populations by AD 900. However, very little research has been carried out in the zone to detect evidence of this proposed climate change. I present results from the first systematic study of wood charcoal from the northwestern frontier, comparing the data from three Classic period ceremonial centers that span the region’s north-south gradient of intensifying aridity. The results indicate that the strongest evidence of environmental degradation is found in the south (where average annual rainfall is the highest), while the sites located farther north (with lower annual rainfall) demonstrate more stable use of wood resources. These findings suggest that anthropogenic impact played a more significant role in regional abandonment than climate change, and that the current models of the process of the collapse in the northwestern frontier may need to be reconsidered.
Full-text available
In this paper, we examine wood charcoal assemblages that were recovered from ash layers in Terminal Classic (A.D. 800-950) burials at the Maya site of Rio Bec to understand the use of fuel wood in funerary rites. Compared to charcoal deposits from domestic and non funerary contexts, the spectrum of wood taxa used in the burial deposits is unique, which suggests specific fire-related practices. Members of the Sapotaceae family and Cordia sp. dominated all contexts and were clearly primary fuels. In contrast, the use of pine (Pinus sp.), which does not grow locally today, was limited to ritual practices. In addition, it seems that a deliberate effort was made to maximize the taxonomic richness of the fuel wood used in burials. Funerary charcoal deposits appear to have been carefully and intentionally "composed" for burning during funerary rites. We propose that this practice materialized the relationship between fire, ash, and the cycles of life and death, which are often symbolized by plant life cycles in the worldview of ancient Maya societies. This study also emphasizes the contribution of anthracological (wood charcoal) analysis to the reconstruction of human behavior and the importance of systematic paleoethnobotanical sampling in funerary contexts.
Full-text available
Palynological archives dating from the Pleistocene-Holocene transition are scarce in the arid zone of the southern Levant. Anthracological remains (the carbonized residues of wood fuel use found in archaeological habitation sites) provide an alternative source of information about past vegetation. This paper discusses new and previously available anthracological datasets retrieved from excavated habitation sites in the southern Levant dating to the Pre-Pottery Neolithic (PPN) period. The available evidence indicates the existence of distinct arboreal floras growing in different ecological niches, which occupied areas that today are either treeless or very sparsely wooded. The anthracological data provide independent confirmation of the hypothesis that early Holocene climate in the southern Levant was significantly moister than at present. Clear North-South and East-West precipitation and associated woodland composition gradients are evidenced. Far from deducing widespread anthropogenic degradation of the regional vegetation, it is suggested that woodland expansion in the semi-arid interiors of the Levant may be attributed to the intensive management of Pistacia woodlands for food, fuel and pasture.
Full-text available
The Uinta Mountains lie along an important transition between major climate regimes and thus are sensitive to major climatic changes and consequent vegetation shifts. Two pollen cores from montane Uinta sites on the Utah–Wyoming border (Marsh Lake Bog and Camp Bog, respectively) were used to analyze vegetation shifts from the latest Pleistocene through the Holocene. The records correspond with those from adjacent portions of the Rocky Mountains, all demonstrating a major shift from alpine steppe to open spruce parkland at ~11,700– 11,200 cal yr BP. A second major shift to lodgepole pine forest began at ~8400 cal yr BP at Marsh Lake Bog and could have been a direct response of the dominant species to warmer conditions and/or the result of subtle changes in the competitive hierarchy of tree species near a sensitive ecotone. The difference in the rate and full expression of lodgepole forest development (taking nearly 2000 years longer at Camp Bog) was probably due to a set of complex and local factors, such as cold air drainage and landscape position. Lodgepole pine forest persisted throughout the late Holocene and appears to be the long-term regional " climax " forest, rather than a seral habitat. Fire does not appear to have triggered the middle Holocene shift from spruce-dominated to lodgepole pine-dominated forests in this area but may have contributed to its long-term persistence to the present day.
With an estimated 10,000 ancient rock art sites, Nine Mile Canyon has long captivated people the world over. The 45-mile-long canyon, dubbed the “World’s Longest Art Gallery,” hosts what is believed to be the largest concentration of rock art in North America. But rock art is only part of the amazing archaeological fabric that scholars have been struggling to explain for more than a century. Jerry D. Spangler takes the reader on a journey into Nine Mile Canyon through the eyes of the generations of archaeologists who have gone there only to leave bewildered by what it all means. The early visitors in the 1890s were determined to recover collections for museums but never much cared to understand the people who left the artifacts. Then came a cadre of young scientists-the first to be trained specifically in archaeology-who found Nine Mile Canyon to be an intriguing laboratory that yielded more questions than answers. Scholars such as Noel Morss, Donald Scott, Julian Steward, John Gillin, and John Otis Brew all left their boot prints there. Today, archaeological research is experiencing another renaissance-a new generation of university-trained archaeologists is determined to unravel the mystery of Nine Mile Canyon using scientific tools and techniques that were unavailable to past generations. Through the words and thoughts of the archaeologists, as well as the more than 150 photos, readers will come to see Nine Mile Canyon as an American treasure unlike any other. As the first book that is devoted exclusively to the archaeology of this unique place, Nine Mile Canyon will evoke fascination among scholars and the general public alike.
Fire is a natural disturbance component and driver of forest composition in the western United States. Cooler/wetter climates are typically associated with less frequent fires and succession of montane forests to more shade-tolerant, fire-sensitive taxa. Native Americans have lived in California since the terminal Pleistocene and used fire to alter the landscape and maximize natural resources; however, determining the extent and impact of anthropogenic burning on California's landscape is difficult because the archaeological record is mostly silent on the subject, and the region's ethnographies mention the practice from a prehistoric context only in passing. Here we show that comparing the prevalence of fire-sensitive to fire-adapted taxa in the pollen record can help distinguish periods when vegetation does not respond as expected to climate change. We argue that the prevalence of shade-intolerant/fireadapted taxa during climatically cool, wet periods such as the Little Ice Age provide evidence for anthropogenic burning. At Holey Meadow, in Sequoia National Forest, we find strong evidence for a Native Americaninfluenced landscape from 750–100 cal yr BP. We also see a strong anthropogenic effect on modern vegetation following European settlement in A.D. 1854, a period marked by a precipitous decline in traditional tribal use of the area and the inception of modern fire exclusion policies. These results indicate that anthropogenic impacts on forest composition can be distinguished from climatic drivers through the use of paleoenvironmental proxies, and further indicate that anthropogenic burning helped structure Late Holocene southern Sierra Nevada biomes.
Previous anthracological research on multi-use pit fills at Kaman-Kalehöyük in central Turkey shows a sudden and dramatic increase in pine use from the Early Iron Age onwards, accompanied by a decline in taxa richness from the Middle Bronze Age to the Late Bronze Age. This paper explores whether fuel choice was the driving factor behind this pattern of change through a comparative taxonomic analysis of wood charcoal assemblages derived from multi-use pits, domestic hearths, and room fills accumulated during the Middle Bronze Age to Late Iron Age occupation at the site. Results from the hearth assemblages parallel those from the pits, documenting a decline in taxa richness from the Middle Bronze Age to the Late Bronze Age, particularly during the Hittite Empire period, followed by a dramatic increase in pine in the Iron Age. However, the room fill composition does not match the same pattern. Results confirm that changes in taxonomic representation represented in the pit fills at Kaman-Kalehöyük was driven by fuel wood selection, with remains from room fills reflecting different selection behaviour relating to construction activities.
Plant material was collected during numerous field tours and expeditions. The geographic origin of the investigated specimens and the species nomenclature are summarized in Table 1. Details are given at Clearly, the largest fraction of plants (about 80%) originated from Europe. Species from outside Europe served mainly to enlarge the taxonomic range and to demonstrate anatomical similarities between tectonically early separated regions.