SDO21078 - The Moors at Arne Coastal Change Project. Geoarchaeological Monitoring and Deposit Modelling

‘Wessex Archaeology (WA) was commissioned by Kier Infrastructure Ltd to undertake a connected programme of geoarchaeological works at the Moors at Arne, Dorset, forming part of the Phase One and subsequent Phase Two works at the site. The geoarchaeological works involved monitoring of Ground Investigations (GI) during summer 2020, including 25 purposive boreholes, followed by a further borehole survey during July 2021, deposit modelling, targeted AMS dating and palaeoenvironmental assessment. The programme of works has been iterative, responding to emerging results and consultation with Historic England to ensure sufficient robust data is available to inform on the geoarchaeological and archaeological potential of the deposits. Geoarchaeological monitoring and deposit modelling of GI logs across Arne Moors has revealed a complex sequence of Pleistocene and Holocene deposits, comprising Pleistocene fluvial sands and gravels, in places overlain by buried soils and a sequence of Holocene peat and alluvium. The Pleistocene deposits represent a high-energy braided river system that was formerly part of the westerly course of the River Solent prior to the Ipswichian, and which at Arne likely date to the Devensian. The deposits form an uneven topography of channels and intervening raised ground that comprises the topographic template on which the Holocene deposits accumulated and which in places persisted as areas of dry ground likely into the Iron Age and perhaps later. Fine-grained deposits were recorded in four of the purposive boreholes (HHWS 209, 217, 220 and 229) and which are hypothesised may represent Late glacial interstadial deposits (c. 14-13 Kya). The topography of the gravel surface is also influenced by a stream valley draining south-north and likely a tributary with the River Frome either within or to the north of the site. A south-north aligned palaeochannel, in the eastern area of the site, and a topographic low in the south-west area of the site, are both likely to be associated with Late Pleistocene and Early Holocene drainage of this stream valley. The Holocene deposits comprise a laterally extensive sequence of peat and alluvium in places up to 5m deep. The peat preserved across the site varies between 0.2 and 4.6 m thick, with the thickest peats preserved in the north-west, south-west and east of the site. In places buried soils are preserved beneath the peat, representing a diachronous former land surface progressively buried by peat formation across the site. Buried soils, particularly in relation to the edges of buried islands are a key context with potential for survival of prehistoric archaeology. Range-finder AMS dating of peat in boreholes HHWS 207, 211, 218, 220 and 223 suggested that the intertidal peat largely accumulated during the Late Mesolithic to Late Bronze Age, punctuated in places by periods of inundation and accumulation of alluvium. Existing radiocarbon dates from the 2018 GI watching brief suggest that localised peat continued to form into the Iron Age and Romano British period in the north-west of the site. Early Mesolithic dates were also produced on peat from borehole HHWS-218 and trench 4 from the southern freshwater area. Preservation of an early Mesolithic peat would be highly significant in view of the scarcity of organic deposits of this date from lowland sites across southern Britain. However, due to concerns that these dates were derived on aquatic material and may include a reservoir effect (producing anomalously old dates), a borehole survey was undertaken during July 2021 to recover additional samples for targeted palaeoenvironmental assessment and radiocarbon dating. Palaeoenvironmental assessment and AMS dating of peat in boreholes HHWS-218 and WA05, supported by deposit modelling, demonstrate that an area of early Mesolithic peat is preserved in the south-west freshwater area of the site. The peat in WA05 began forming between 9120–8640 cal. BC, with the hazel-rise dated to between 8540–8300 cal. BC. The deposit modelling suggests that this peat was forming within a depression in the underlying Pleistocene topography in the south west of the site, although somewhat irregular in shape but in the order of 400m north-south and east west. It is unclear if this low in the gravel surface is part of the former Pleistocene drainage, which may be related to a south-north draining valley and which could be further connected with a defined north-south channel identified along the eastern half of the site. The remaining peat across the site formed in response to fluctuating sea-level tendencies. These coastal peats would have represented a range of environments from reed swamp, sedge-fen to wet woodland. AMS dates on these peats are derived largely on fragments of common reed, and although every effort was made to date the aerial component of the plant, it cannot be ruled out that fragments of root were present in some samples, producing anomalously young ages. However, the dates on these peat correspond well with our understanding of peat formation in coastal contexts across southern Britain during the late Mesolithic to Iron Age. This peat gradually encroached across the site, incorporating the area of early Mesolithic peat in the south-west freshwater area, and progressively encroaching on areas of dry ground in the centre of the site, some of which may still have been connected to dry land surrounding the site during the Neolithic. Peat formation ceased across much of the site in the Bronze Age, succeeded by deposition of silty-clays representing intertidal saltmarsh and mudflat environments. Localised areas of peat persisted in the western margins of the site with islands of dry ground persisting in the centre of the site – archaeological evidence for activity during the Iron Age on these islands may be related to exploitation of the intertidal saltmarsh for stock grazing. The deposits preserved across the site represent a valuable and regionally significant resource informing on past physical landscape, vegetation, climate and environmental change in a lowland coastal setting during the Late Pleistocene and Holocene. Suggestions are made for further works to assess the preservation potential of key palaeoenvironmental remains and their potential to contribute to the aims of the project, as outlined in a series of Specific Research Objectives (SROs). Further work would focus on targeted retained sequences, including fine-grained deposits preserved in fluvial sands and gravels and key sequences of peat and alluvium. Fine-grained deposits preserved in Pleistocene fluvial sands and gravels may date to warmer interstadial periods during the last ice age and would represent a significant geoarchaeological resource on Late Pleistocene physical landscapes and environment. The sequence of Holocene peats and alluvium are a regionally important resource for understanding post-glacial landscapes, environment over a period of several thousand years, including evidence for human activity/land-use in the context of changing climates, sea-level rise and land-sea interactions.’