Luminescence Dating of Windblown Sands from Archaeological Sites in Shetland (Early Modern Farmstead, Broo Peninsula; Norse settlement, Sandwick South, Unst)

Kinnaird, T. C., Sanderson, D. C.W. and Bigelow, G. (2014) Luminescence Dating of Windblown Sands from Archaeological Sites in Shetland (Early Modern Farmstead, Broo Peninsula; Norse settlement, Sandwick South, Unst). Technical Report. SUERC, University of Glasgow.

110521.pdf - Published Version



This report provides optically stimulated luminescence (OSL) results for key stratigraphies from the vicinity of a Norse house at Sandwick South (Unst), and from a farmstead at Broo, Quendale (Mainland Shetland). At both sites, the sediment sequences associated with the nearby structures record accumulations of blown-sand, potentially linked to climatic instability. David Sanderson and Gerry Bigelow visited the sites in November 2013 to sample undisturbed sand sequences associated with the built structures for OSL dating. Sampling pits were opened at both sites: at Sandwick South, the 2.1 m deep trench was positioned WSW of the Norse House; at Broo, the 1.2 m deep trench was positioned E of the main building complex, within the farmyard enclosure. During fieldwork portable OSL equipment, in combination with field spectrometry, was used to appraise luminescence stratigraphies, and identify the key units for OSL dating. The test trench at Sandwick South was strategically positioned to examine the midden sequence in proximity to the built structure: profiling samples were collected from the lowest sands (3 samples), and from sand sequences enclosing the various midden units (7 samples), including clean, post-abandonment sands (4 samples). The field profile showed an overall increase in luminescence signals with depth, consistent with an age-depth progression. Furthermore, each of the stratigraphic breaks identified in the field, corresponded with a step in signal level, suggesting temporal breaks across the sedimentary boundaries. Four OSL dating samples were collected, along with in-situ gamma spectra to cover the key stratigraphic units. Field gamma dose rates were below 0.1 mGy a-1, implying that low dose rates would be encountered from the dating samples. This is consistent with the local lithology, which is also quartz poor. Nonetheless the presence of stratigraphically progressive luminescence signals in profiling confirmed that the material contained a luminescence phase with dating potential. Laboratory profiling confirmed that luminescence sensitivities were also low in both quartz and polymineral fractions. Quartz was present in the sediments, but when examined had very low luminescence sensitivity, which combined with low dose rates, precluded application of the quartz SAR method. Electron microscopy confirmed the presence of a high sodium feldspar fraction, which was separated and used for infra-red stimulated luminescence (IRSL) dating. Dose estimates were undertaken using an adapted SARA protocol, incorporating long overnight preheats before first measurement, to mitigate short-term fading effects (following the suggestions of Sanderson, 1988a). To accommodate the low dose rates the sediment samples were light protected and subjected to prolonged high resolution gamma spectrometry prior to mineral extraction. Thick source beta counting performed on subsamples were also conducted for extended periods. Internal alpha dose rates have been estimated on the basis of ICPMS analyses of comparable feldspars, and make a minor (~10%) contribution to the overall dose rates from this material (typically 0.4-0.5 mGy a-1). Fading tests have been initiated, but so far no corrections have been applied. Using these procedures, the following chronology is suggested for the Sandwick South site: the lowest sand in the position of the test pit is dated to AD 1220 ± 120 (SUTL2603), and the mixed sand/midden dated to AD 1290 ± 80 (SUTL2602) and AD1210 ± 80 (SUTL2601). These three age estimates are indistinguishable within error, implying a rapid sediment accumulation. If they are treated as synchronous the combined result is AD1240 ± 50. The overlying sand in this profile gives a result of AD 1770 ± 50. At Broo, east of the main building complex and in the main farmyard enclosure, the pit provided access to undisturbed sands in contact with substrata, and to sands beneath and above a flagstone-lined drain, thereby providing TPQ and TAQ for the period of construction of the drain, and the overlying sequence of sands leading to the upper, post-abandonment sands. The field profile again showed a progression in luminescence signals with depth, with steps in signal level across the main stratigraphic breaks. Importantly, the lowest sands, which directly seal the substrate, are characterised by signal intensities an order of magnitude larger than those in the uppermost sands. As with earlier analyses from Broo, it was possible to apply the quartz SAR technique. Laboratory profiling was conducted, which verified the presence of both quartz and feldspars and corroborated the field profiling results. The quartz OSL SAR sand-based chronology obtained for the section at Broo encompasses a span from the late 14th century through to the early 18th century. The lowest sand in the section is dated at AD 1370 ± 40 (SUTL2608). The flagstone-lined drain was constructed in the period between AD 1510 ± 60 (SUTL2607) and AD 1550 ± 60 (SUTL2606). The clean sands identified in the upper part of the profile are dated to AD 1710 ± 60. In addition to these formal samples, two samples were also dated from a test trench cut immediately NW of the building to examine a midden sequence outside the main enclosure. The section was sampled in 2012 by Zoe Outram (University of Bradford), in the absence of field spectrometry or OSL profiling equipment. In addition to the two dating samples above and beneath the midden sequence, a series of 20 bulk gamma spectrometry samples was collected to permit retrospective analysis of the external gamma dose rates. The gamma dose rates were reconstructed from 17 of these giving wet gamma dose rates of 0.90 and 0.87 mGy a-1, which are comparable with the measured values recorded in the adjacent pit. Individual quartz OSL SAR ages fall in 18th century AD, implying a short period of accumulation (individual ages are AD 1720 ± 15 and AD 1780 ± 35, SUTL2576 and 2577, respectively). Interestingly both sites show periods of significant sands accumulation in the 18th century. At Sandwick South the 12th/13th century activity falls within the period of radiocarbon dates from the site. There may be scope for further refinement if the internal activity were determined directly, and also data reviewed following a prolonged fading test. At Broo, the lowest sand extends the chronology within the vicinity of the settlement, with early 16th century dates above and below the drain corroborating an earlier determination of 1540 ± 40AD (SUTL 2441) obtained in the vicinity of the Broo II site. These new determinations thus add to the growing body of luminescence dates for windblown sands in Shetland.

Item Type:Research Reports or Papers (Technical Report)
Glasgow Author(s) Enlighten ID:Kinnaird, Dr Timothy and Sanderson, Professor David
Authors: Kinnaird, T. C., Sanderson, D. C.W., and Bigelow, G.
Subjects:G Geography. Anthropology. Recreation > GB Physical geography
College/School:College of Science and Engineering > Scottish Universities Environmental Research Centre
Publisher:SUERC, University of Glasgow
Copyright Holders:Copyright © 2014 The Authors
Publisher Policy:Reproduced with the permission of the authors.

University Staff: Request a correction | Enlighten Editors: Update this record