|
(To view an image, click on it;
|
Preliminary
2003 Results! |
Conclusions |
Nearest town: Bismarck; State/country: North Dakota; Type of site: Prehistoric fortified earth lodge village (ancestral Mandan); Date of site: uncertain; late 1400s - mid-1780s AD; Collaborators: PaleoCultural Research Group, Flagstaff, AZ, State Historical Society of North Dakota, University of Missouri.
The Double
Ditch site (32BL8), located on a high terrace above the Missouri River
immediately north of present day Bismarck, is preserved within a state
park, the Double Ditch State Historic Site. When Lewis and Clark passed
by in 1804 it had probably been abandoned by the Mandan for a couple of
decades, most likely due to a smallpox outbreak in the 1780s. It has always
been understood that Double Ditch is an unusually large site. At approximately
4.8 ha within the second or outer ditch seen in the topography, it is
one of the largest fortified settlements along the Middle Missouri River.
It is also distinctive in the size and number of its large midden mounds,
some over 3 m, the many shallow depressions 12-18 m in diameter that signify
the loci of former houses, and its readily apparent double fortification
system from which it derives it's name.
A small exploratory geophysical survey was undertaken at this site in 1997 that indicated good subsurface contrasts and anomaly definition in magnetometry and resistivity data (Kvamme 2000). With expected increases in tourism, the upcoming Lewis and Clark bicentennial (in 2004), and recent improvements to the property in the form of a parking area and signage, large-scale geophysical surveys were undertaken in 2001 (Kvamme & Kvamme 2001; Kvamme 2002). Further geophysical work was again carried out in 2002 to expand the areal coverage of the surveys, the focus of this paper. In each year, subsequent to the geophysical surveys, archaeological fieldwork was conducted by Stanley A. Ahler of the Paleo Cultural Research Group (PCRG), Fern Swenson of State Historical Society of North Dakota (SHSND), and W. Raymond Wood leading the University of Missouri archaeological field school. Hand corers were employed in 2001 (Ahler & Swenson 2001) and larger excavations in 2002 (Ahler et al. 2002). These archaeological testing programs were largely guided by the geophysical results, but their findings, in turn, gave insights about the meaning of the geophysical signatures, helping us to understand what we are seeing in the data.
Magnetic
gradiometry survey of an additional 68 grids (20x20 m) was accomplished
in an attempt to define the outer fortification system. This proved challenging
because in the outer limits of the site, features of the trenches and
bastions yield contrasts of only .4-.6 nT, approaching the noise limits
of our instrument! The survey was successful, defining many new trenches,
double trenches, bastions and unanticipated (and currently unexplained)
features.
Dr.
Tommy Ike Hailey (Northwestern State University, Natchitoches, LA) joined
forces this year, bringing his powered parachute! We obtained wonderful
color aerial imagery and made use of the Center
for Advanced Spatial Technologies' new thermal camera (a Ratheon Palm
IR). Results are shockingly good, revealing archaeological features with
a clarity not anticipated and leading to the discoivery of a host of new
features. The image at left shows a large portion of the central village
core, including portions of the two inner fortification ditches, many
earthlodges, and a large borrow pit. It measures approximately 140 m across.
Dr. Hailey is currently working with these data which we expect will bring
many new insights and directions to this project.
Stay tuned for future updates!
Type of survey: Magnetic gradiometry; Instrument: Geoscan Research FM-36 fluxgate gradiometer; Prospection depth: up to 1.5 m; Sampling interval: 0.5 x 0.5 m; Area surveyed: 17,200 m2; Date of survey: 6/01.
Comments: The magnetic survey sampled the known site area in a
series of transects (left figure), and yielded a number of new insights.
(1) Numerous 1-1.5 m diameter circular anomalies (2nd figure), distributed
largely outside of houses, were mapped that mostly represent storage pits
(as shown by testing, and by analogy to results at other similar sites,
like Huff Village [Ahler & Kvamme 2000]). (2) A number of rectangular-shaped
houses were better defined (beyond their visible surface depressions)
with such interior features as central and auxiliary hearths and storage
pits revealed in the data (3rd figure). (3) Two very large circular borrow
pit areas were made clear, measuring 32 and 50 m in diameter, that are
devoid of anomalies (4th figure). Soil coring (Ahler & Swenson 2001)
showed the A-horizon is completely missing in these areas. They are interpreted
as places where soil was mined, possibly for earthlodge construction,
but another theory is later advanced based on the 2002 findings. (4) A
number of linear features (last figure) were revealed in the data that
coring suggested were shallow ditch features, but their function could
not be explained (Ahler & Swenson 2001; Kvamme 2002).
Type of survey: Resistivity; Instrument: Geoscan Research RM-15 with MPX-15 multiplexer; Prospection depth: 1.0 m; Sampling interval: 1.0 x 1.0 m; Area surveyed: 17,200 m2; Date of survey: 6/01.
Comments: The resistivity data gathered in 2001 were less insightful
than the magnetometry data, but nevertheless produced several important
results (left figure). The data were sampled primarily at a target depth
of 1 m owing to our initial assumption of a long occupation (suggested
by the size of the midden mounds) that would suggest more deeply buried
features (the archaeological coring work performed a short time later
showed the reverse to be true, with the former occupation surface lying
only about 50 cm deep [Ahler & Swenson 2001]). The resistivity data
tend to indicate midden mounds as high resistivity features, probably
owing to their generally increased dryness that is due, in part, to a
topographic effect (high points tend to be drier), but also to decreased
compaction and their being a focus of intensive rodent activity (larger
pore spaces). The data also point to some high resistivity features, quite
possibly dispersed middens that are not visible as mounds on the surface.
Another phenomenon that the resistivity data tend to emphasize is a zone
of high resistivity around the perimeter of many of the obvious house
depressions (right figure). They are interpreted as zones of "roof
melt," eroded sediments from earthlodge roofs that may exhibit high
resistance owing to compaction differences and the topographic effect
(Kvamme 2002).
2001
Summary. The view of Double Ditch in 2001 was much richer thanks
to the geophysical work, with many more details known about features and
feature distributions within the site, and a better realization of the
fantastic density of archaeological remains this village must contain.
Yet, in an overall sense, our perception of Double Ditch was little changed:
it remained an atypically large settlement with a secondary fortification
system containing unusually large midden mounds
The geophysical work in 2002 will change forever the way we view Double Ditch and, indeed, certain aspects of northern Great Plains prehistory (Kvamme and Kvamme 2002). In 2002 the goals of the geophysical work were to greatly enlarge the areal coverage of the surveys, particularly in the northeastern section and the village core, to place a new eastern transect to explore that region of the site, and to attempt to define former excavation trenches placed by Will and Spinden in 1905 in order that archaeological teams led by Ahler et al. (2002) could re-investigate them. All these goals were accomplished with an additional 2.1 ha of magnetic survey, 1.9 ha of resistivity, and a more limited amount of additional GPR work.
Type of survey: Magnetic gradiometry; Instrument: Geoscan Research FM-36 fluxgate gradiometer; Prospection depth: up to 1.5 m; Sampling interval: 0.5 x 0.5 m; Cumulative area surveyed: 38,000 m2; Date of survey: 6/01 & 6/02.
The most significant and obvious findings in the 2002 results is not one,
but two additional fortification systems that surround the village.
Of equal importance are the presences of systematically placed bastions
with forms that point to a much greater antiquity of the site than previously
thought. Six additional points related to these discoveries are emphasized.
We
speculate that these mounds, while undoubtedly composed of midden materials
(trash and refuse), were purposely sited as integral components of the
fortification systems. Placed immediately inside the fortification ditches
their steep outer sides form a natural defense (perhaps even a rampart)
;
their heights may have allowed them to serve as observation platforms
within the defensive network. It is further speculated that a purpose
of the large borrow pit (see above, and right figure), located only
25-80 m away from these works, may have been to contribute to the construction
of these mounds.
The 2002 magnetic data also offer further evidence of rectangular houses,
including several that are closely packed together with indications of
large numbers of associated storage pit features (left figure). Of equal
significance, we believe there is good evidence in the data of circular
earthlodges as well (right figure). These findings support the inference
of a long occupation from perhaps the 15th century to immediately before
the time of Lewis and Clark, but we would like to see validation through
archaeological excavation.
Type of survey: Resistivity; Instrument: Geoscan Research RM-15 with MPX-15 multiplexer; Prospection depth: .5 m; Sampling interval: .5 x .5 m; Cumulative area surveyed: 36,000 m2; Date of survey: 6/01 & 6/02.
The resistivity survey, now totaling 3.6 ha, was performed with a target
depth of 50 cm to better image the many features that we now know are
closer to that depth. While an incongruity results when these data are
matched with the 1 m data from 2001 (left figure), the new information
provides much better detail that reveals what are probably smaller midden
features, and perhaps some interior details within houses. Possibly owing
to the unusually hot and dry summer or to inherent soil properties, resistivity
contrasts were somewhat poor or even non-existent in the vicinity of the
newly discovered fortification systems (right figure), but provide important
evidence that corroborates the magnetic findings. A puzzling result is
that certain portions of these ditches exhibit a more common low resistivity,
while other segments indicate high resistivity. This phenomenon cannot
be explained at this time, but perhaps indicates differences in the nature
of their fills.
One
goal of the geophysical project was to attempt to better reveal the excavations
made by Will and Spinden (1906) in order that they could be re-opened
by the archaeological teams led by Ahler et al. (2002). Resistivity was
particularly successful in this regard; the unusual triangle of trenches
and other holes they placed over a house and other features in one region
are clearly visible in the data.
The geophysical results at Double Ditch further emphasize the important role that these methods can play in archaeological projects. They work well as feature discovery tools by locating and mapping anomalies of potential archaeological significance. We believe that geophysics offers much more, however. Clear imagery in many cases can illustrate interpretable anomalies that do not require testing for their identification. In the present case, the fortification ditches, bastions, some of the houses, hearths, storage pits, and previous excavation holes can be understood in the imagery. While it is true that some of these features were initially unrecognizable and uninterpretable only a few years ago, in particular the storage pit features, experience gained through archaeological testing (Ahler & Kvamme 2000; Ahler & Swenson 2001; Ahler et al. 2002) now enables their more ready recognition.
The ability to identify archaeological elements in geophysical data, coupled with subsurface mappings of vast areas that modern technology allows, means that we can now view buried archaeological features throughout large regions, and better interpret them by recognizing their context and associations with other features. This capability potentially opens the door to a new kind of landscape archaeology where our perceptions about the past are not confined to the few square meters common to archaeological excavations, but to areas of many hectares. It also illustrates the necessity of surveying large contiguous regions to better understand the meaning of anomalies.
At Double Ditch, thanks to geophysical surveys, we now know we have a
truly gigantic settlement of much greater complexity, time depth, and
significance. We hope that additional geophysical work will be possible
in the future to completely map this settlement and it's defenses, and
to discover yet more fortifications about this world class site.
Back
to "LIBRARY OF GEOPHYSICAL IMAGERY" home page
(updated: 8/03)