close

Вход

Забыли?

вход по аккаунту

?

Ground-penetrating radar time-slices from North Ballachulish Moss.

код для вставкиСкачать
Archaeological Prospection
Archaeol. Prospect. 11, 65–75 (2004)
Published online 12 May 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/arp.223
Ground-penetrating Radar Time-slices
from North Ballachulish Moss
ERICA UTSI*
Utsi Electronics Ltd,1School Lane, Aldreth, Ely CB6 3PL, UK
ABSTRACT
In the late1990s ground-penetratingradar (GPR) wasused to mapthe underlying topographyof North
Ballachulish Moss aspart ofan archaeologicalevaluation ofthe area that wasunder threat ofdevelopment.In the process a series of radar anomalies were identified.These were confirmed subsequently
by sediment coring and trial excavation to correspond to a buried prehistoric surface with associated
man-made artefacts.These GPR data have now been used to develop horizontal time-slices.The series of time-slices is presented showing the development of discrete surfaces and their relationship to
an adjacent headland.The orientation of the site andits proximity to the location of a buried prehistoric
wooden figure suggest ritual importance. Although comparable data in similar locations are lacking,
the possibility exists to shed further light on the Ballachulish evidence without destroying wetland resourcesiffurther GPRinvestigationcanbecarriedoutonother wetlandsitesofrecognizedritualsignificance.Copyright 2004 JohnWiley & Sons,Ltd.
Key words: peat; radar; prehistoric surfaces; time-slices; prehistoric ritual and wetlands
Background
North Ballachulish Moss (Figure 1) has been
known to be a wetland area of archaeological
importance since the late nineteenth century
when both organic and inorganic remains were
recovered from the peat. These include a circular
wattled structure, flints, cists and casks of bog
butter. By far the most spectacular was the discovery of a near life-size wooden female figure
dated to the first millenium BC buried within the
peat in a manner reminiscent of the Danish bog
bodies of the same period (Christison, 1881;
Coles, 1990).
More recent archaeological assessment of the
site confirmed that, despite continuing attrition,
areas of deep peat remained within the southern
* Correspondence to: E. Utsi, Utsi Electronics Ltd, 1 School
Lane, Aldreth, Ely CB6 3PL, UK.
E-mail: erica.utsi@utsielectronics.co.uk
Copyright # 2004 John Wiley & Sons, Ltd.
portion of the Moss (Figure 2). Recognition was
given to the potential for preservation of organic
remains and, because of the site location, the
possibility of Mesolithic artefacts sealed in the
subpeat layers (Pollard, 1993).
Owing to its Highland location and the scarcity of alternative sites the Moss has come under
increasing threat of use for housing development. In 1996 the Centre for Field Archaeology,
University of Edinburgh (CFA), funded by Historic Scotland, carried out an archaeological and
palaeoenvironmental evaluation of the southern
portion of North Ballachulish Moss. As part of
this study, CFA commissioned a ground-penetrating radar (GPR) survey of approximately half
of the southern portion of the Moss in order to
map the contours of the peat basin and to detect,
if possible, any archaeological features within the
peat (Utsi Electronics, 1996). The GPR survey
was extended into the remainder of the southern
area of the Moss in 1998 (Utsi Electronics, 1998;
Clarke et al., 1999).
Received 14 February 2003
Accepted 21 January 2004
66
E. Utsi
Figure1. Location Map (Clarke et al.,1999).
The GPR surveys and the
environmental evidence
Although associated finds from the previous
century implied that there was a reasonable
Copyright # 2004 John Wiley & Sons, Ltd.
probability of archaeological remains being present in the Moss, it was not known what form
these might be expected to take nor if they would
be identifiable by GPR. The original (1996) investigation was thus essentially an exploratory
Archaeol. Prospect. 11, 65–75 (2004)
GPR Survey of North Ballachulish Moss
67
Figure 2. North Ballachulish Moss from the South.
survey. The survey strategy combined total site
coverage at 10-m intervals to define the peat
contours with a more detailed investigation of
one area at 2-m intervals (‘Area 4’, Utsi Electronics, 1996). This area, 50 m by 60 m, was selected
during the survey to cover the part of the Moss
where anomalous signals had been returned
from within the peat.
Data collected in this first GPR survey was
presented to the CFA both as radar profiles and
as contoured maps of three distinct peat basins
(Utsi Electronics, 1996; Clarke et al., 1999). All of
the major anomalies were located within the
deepest of these basins. Despite the relatively
large transect spacing of 2 m in Area 4, continuity
of the structures within the peat was suggested
by several of the adjacent profiles: compare
Figures 3 and 4.
The major anomalies were plotted on to the
contour map of the peat basins: this suggested
that the anomalies formed a series of distinct
groups (Figure 5). This method of presentation
did not allow for the depth variation visible in
the profiles (Figures 3 and 4). This paper presents
the data from Area 4 in time-slices in order to
make good that omission.
Copyright # 2004 John Wiley & Sons, Ltd.
Following the first GPR survey, trial excavation and test coring of the site were carried out in
order to determine whether or not the anomalous
signals represented past human activity. The
position of the 4 m 4 m trench is shown in
Figure 5. This revealed the edge of a black oily
compacted clay platform and some associated
worked wooden pegs. The clay matrix was
packed with fragments of quartz, quartz-rich
sandstone and stones. Radiocarbon dating of
the associated peat and the worked pegs suggested a date in the first to second millenium BC
(Clarke and Stoneman, 1998).
Macrofossil and microfossil analyses from the
site suggested a major change in vegetation,
probably owing to the impact of human activity
at ca. 110 cm depth. Woodland clearance was
indicated from the third millenium BC onwards
and the environmental effects of human activity
in the vicinity of the site persisted in the pollen
record to the present day. The environmental
evidence was also consistent with a relatively
dry bog surface at the time when the platform
was constructed (Clarke and Stoneman, 1998).
Two years later, the remaining half of the Moss
previously excluded from GPR survey was
Archaeol. Prospect. 11, 65–75 (2004)
E. Utsi
68
Figure 3. Profile 48.
Figure 4. Profile 49.
examined. As this second survey did not reveal
any additional areas of deep peat or major anomaly groups beyond those already discovered, no
further detailed investigations were carried out
(Utsi Electronics, 1998).
The equipment used for both GPR surveys was
a Groundvue 2, a low-frequency broadband GPR
with a range of 30 MHz to 100 MHz. The radar
used for the 1996 survey had horn antennae: the
one used for the second survey used twin
arrayed bowtie antennae.
Copyright # 2004 John Wiley & Sons, Ltd.
Time slices
As previously explained, site coverage was not
initially determined with a view to constructing
time-slices. Ideally the data sets for time-slices
should be derived from closely spaced parallel
profiles. Transect spacing should be less than one
half the wavelength of possible reflections returned
from the smallest target to be mapped (Conyers
and Goodman, 1997). Assuming a 50 MHz central
frequency for the radars used, this implies a
Archaeol. Prospect. 11, 65–75 (2004)
GPR Survey of North Ballachulish Moss
69
Figure 5. Peat depth and radar anomaly map resulting from the1996 and1998 GPR Survey (Clarke et al.,1999).
Copyright # 2004 John Wiley & Sons, Ltd.
Archaeol. Prospect. 11, 65–75 (2004)
E. Utsi
70
Figure 6. Outline of the peat basin at126.3 ns (2.22 m).
transect spacing of 0.5 m. In theory this should
preclude the use of time-slicing on these data.
The effective target area is measured by the
radius of the Fresnel zone. This can be calculated
by the formula
p
zone diameter ¼ 2 ½ðd þ =4Þ2 d2 where d is target depth and is the centre
frequency wavelength generated by the radar.
For a Groundvue 2, in a waterlogged medium of
dielectric permittivity of 73 (as measured by the
WARR method during the survey), this gives the
following theoretical target areas:
depth
0.5 m 1 m
2m 3m 4m
zone diameter 1 m 1.2 m 1.8 m 2 m 2.4 m
As the anomalies lie within 2-m of the presentday surface, the resolution of time-slices constructed from profiles taken at 2-m intervals
will be less than optimum.
However pioneering work on GPR time slicing
(e.g. Goodman and Nishimura, 1993) has shown
that it is possible to obtain useful patterning from
Copyright # 2004 John Wiley & Sons, Ltd.
time-slices constructed from relatively widely
spaced two-dimensional profiles even though
the resolution capability is reduced.
The data from Area 4 were combined into a
three-dimensional block and time-slices generated. The results are visible in Figures 6–11. In
these time-slices the x axis represents the 1996
baseline and the y axis runs from the baseline to
the 50 m line (see Figure 5).
Discussion
The sequence of time-slices illustrates the development of a series of four large platforms within
the peat basin. Three of these cluster around a
prominent headland in a manner which suggests
that this itself may form part of the complex of
structures. The fourth lies more centrally in the
deepest part of the moss and approximately due
west from the headland.
The identification of these anomalous structures with clay platforms is reasonable, given the
results of the trial excavation. The excavation
trench lies between the two larger platforms
Archaeol. Prospect. 11, 65–75 (2004)
GPR Survey of North Ballachulish Moss
71
Figure 7. Lower edge of prehistoric surfaces at101.3 ns (1.78 m).
Figure 8. Platform development at 95 ns (1.67 m).
Copyright # 2004 John Wiley & Sons, Ltd.
Archaeol. Prospect. 11, 65–75 (2004)
72
E. Utsi
Figure 9. Platform development at 85 ns (1.49 m).
Figure10. Possible Upper Surfaces at 75 ns (1.32 m).
Copyright # 2004 John Wiley & Sons, Ltd.
Archaeol. Prospect. 11, 65–75 (2004)
GPR Survey of North Ballachulish Moss
73
Figure11. Possible Upper Surfaces at 66.25 ns (1.16 m).
close to the headland (cf Figures 5 and 10) and
the results are consistent with finding the western edge of the most easterly platform close to
the headland. The site at North Ballachulish was
waterlogged: dielectric permittivity measured
during the survey by the WARR method was
73. This makes it surprising to receive such
strong signal responses from clay contained
within the peat. In the absence of evidence to
the contrary, it must be assumed that either the
compaction of the clay is sufficiently great or that
the inclusions are sufficiently plentiful for the
platforms to respond electrically in a manner
similar to building materials. There is no significant signal attenuation at these depths.
The size of the structures is remarkable: the
smallest is of the order of 4 m by 5 m; the largest
approximately 5.5 m by 7 m. It is, however, possible that the discontinuities visible in the timeslices may indicate subsidiary structures that are
not fully defined because of the large transect
spacing.
Dating of the site depends upon the environmental work and radiocarbon dating both of the
peat and of the worked wooden pegs. The marks
Copyright # 2004 John Wiley & Sons, Ltd.
on the latter indicate the use of metal tools
(Clarke and Stoneman, 1998). It is possible that
the site contains some phasing of activity. Given
that the peat formation is a slow process (dates
from above and below the excavated platform
differ by the order of 1900 years) and that the
ground surface of the Moss appears to have been
relatively solid when the platform was laid, it is
reasonable to expect that the prehistoric surface
would be level or nearly so.
Comparison of Figures 7, 9 and 11 suggests
that two of the platforms lie slightly lower in the
Moss: the smallest platform adjacent to the headland and the one lying in the centre of the basin.
On the basis of GPR evidence alone, it is not safe
to assume that this represents dating differences
because this could equally well represent the
result of other causes such as post-depositional
damage, structural differences or relative sinking
of the features within the peat.
What, if anything, do these time-slices tell us
about the nature and use of the site? As a
technique, GPR is notoriously limited as a tool
of site interpretation and therefore any conclusions drawn must be treated as conjecture. The
Archaeol. Prospect. 11, 65–75 (2004)
E. Utsi
74
platform in the centre of the peat basin, to the
west of the headland, is different in orientation to
the other platforms. Its distance from the sides of
the peat basin is much greater. This raises the
possibility that its purpose was also different
from that of the platforms closer to the headland.
It also suggests that the centre of the Moss was
important to the people who constructed and
used the site.
The headland itself rises steeply. At the time
when the platforms were constructed, this would
have been a prominent feature of the surrounding landscape. The clustering of the three adjacent platforms suggests that the headland itself
may form an important aspect of this site.
Although the possibility that this represents a
habitation site cannot be ruled out, there is some
evidence to suggest use for a ritual purpose. The
site’s principal axis of symmetry from the platform in the centre of the Moss to the headland is
orientated east to west. The relatively dry conditions prevalent on the Moss surface at the time
that the platforms were constructed lessens the
likelihood that a central position in the peat basin
was chosen in order to afford greater protection
to people, animals or food stores. It is also interesting to note that the wooden figurine referred
to in the introduction was found face down at the
lowest level of a nearby area of deep peat, pinned
down by a hurdle and possibly surrounded by a
wooden structure from which ‘several pole-like
sticks’ remained (Christison, 1881). Peat depth
evidence from the GPR survey indicates that this
basin would have been the nearest deep peat
deposit (cf. either Christison (1881) or Pollard
(1993) with Figure 5).
Comparison of Figure 3 with the time-slice
series reveals a small number of anomalies on
the periphery of the peat basin that do not
correspond to the platforms in the time-slices.
By and large these do not exhibit any continuity
in depth: some lie at the bottom of the peat and
others at different levels within the peat.
archaeological importance (Pollard, 1993; Clarke
et al., 1999). The importance of managing the site
both for its value as part of a rapidly diminishing
wetland resource and for its potential for archaeological preservation has been discussed elsewhere (Pollard, 1993; Coles and Coles, 1995).
Although it is not possible to define the use
that prehistoric people made of the site without
further investigation, the time-slices are consistent with the possibility of ritual importance.
Much of the UK’s wetland resources are vanishing or under threat. At Ballachulish Moss the
remaining limited deep peat deposits continue to
enjoy the protection of waterlogging. Further
investigation would effectively destroy the remaining deep peat. In these circumstances it is questionable whether further excavation of the site can
be justified. A GPR survey has the merit of being
capable of non-intrusive monitoring of wetlands
with low electrical conductivity. It is probable
that the spatial definition of the Ballachulish site
could be improved if the GPR survey were to
be repeated at closer transect spacing. Although
our understanding of the purpose and prehistoric
activity that it represents might be less than that
obtained from a full scale excavation, the potential
for gaining comparative information from other
wetland sites should not be ignored.
It is important that serious consideration is
given to carrying out GPR surveys particularly
in areas where there is good reason to suspect
the presence of archaeological remains. If this is
done it should be possible to build up a data
base of managed sites that, by comparison, may
reveal more information about their nature
and use.
The parallels between the Ballachulish figure
and the bog bodies of northern Europe have
been described elsewhere and there is good
reason to associate these (Coles, 1990). The
obvious comparison for the Ballachulish GPR
data lies in the wetland sites associated with
finds of this nature, assuming that sufficient
peat deposits remain.
Conclusions
The series of time-slices derived from the GPR
survey of North Ballachulish Moss confirms previous observations that this is a site of immense
Copyright # 2004 John Wiley & Sons, Ltd.
Acknowledgements
The author would like to thank Dr Ciara Clarke
and Historic Scotland (particularly Sarah Govan
Archaeol. Prospect. 11, 65–75 (2004)
GPR Survey of North Ballachulish Moss
and Richard Hingley) for commissioning the
original GPR survey and for subsequent permission to use the data for this paper.
References
Clarke CM, Stoneman R. 1998. Archaeological and
palaeoenvironmental investigations at North
Ballachulish Moss, Highland, Scotland. In Recent
Developments in Wetland Research, Raftery B,
Hickey J (eds). Proceedings of a Conference
held by the Department of Archaeology, University College Dublin and the Wetland Archaeology Research Project (WARP).
Clarke CM, Utsi E, Utsi V. 1999. Ground Penetrating Radar Investigations at North Ballachulish
Moss, Highland, Scotland. Archaeological Prospection 6: 107–121.
Christison R. 1881. On an ancient wooden image,
found in November last at Ballachulish Peat
Moss. Proceedings of the Society of Antiquaries of
Scotland 15: 158–178.
Copyright # 2004 John Wiley & Sons, Ltd.
75
Coles B. 1990. Anthropomorphic wooden figures
from Britain and Ireland. Proceedings of the Prehistoric Society 56: 315–333.
Coles J, Coles B. 1995. Enlarging the past. The
contribution of wetland archaeology. Society
of Antiquaries of Scotland Monograph Series
11: 132 and 154–159.
Conyers LB, Goodman D. 1997. Ground-penetrating
Radar: An Introduction for Archaeologists. Altamira
Press: London; 153–154.
Goodman D, Nishimura Y. 1993. A ground radar
view of Japanese burial mounds. Antiquity 67:
349–354.
Pollard A. 1993. Ballachulish Moss Assessment. Report No. 136, Glasgow University Archaeological
Research Division, Glasgow.
Utsi Electronics. 1996. Ground Radar Survey of a
site within North Ballachulish. In CFA Data Structure Report No. 293. Centre for Field Archaeology:
Edinburgh.
Utsi Electronics. 1998. Ground Radar Survey of a Site
within North Ballachulish. Report for the Centre for
Field Archaeology: Edinburgh.
Archaeol. Prospect. 11, 65–75 (2004)
Документ
Категория
Без категории
Просмотров
0
Размер файла
598 Кб
Теги
grounds, north, penetration, times, radar, ballachulish, slices, moss
1/--страниц
Пожаловаться на содержимое документа