Archaeology in the Severn Estuary 14

Although the presence of these submerged prehistoric woodlands have been observed in a number of studies, rarely have they been recorded and analysed in the detail they deserve. Submerged forests are after all exceptionally well preserved remnants of the original woodland offering high palaeoecological resource potential. Earlier studies have attempted to reconstruct these environments on the basis of one main analytical approach, mainly pollen analysis, less commonly wood identification, seeds and beetles. This study attempts to utilise the full range of those resources by using a multiproxy approach to reconstruct part of the extensive submerged forest at Goldcliff East.

Extensive fieldwork and systematic sampling over the past three years has produced the most complete data set of palaeoentomological information derived from any single estuary. Analysis of these samples makes it possible to infer that, over varying spatial and temporal scales, the past biomes of the Severn Estuary (Figure 1) were a complex mosaic of habitats from saltmarsh to reed swamp and raised bog, a superb illustration of haloseral and subsequent hydroseral transition. Despite a plethora of evidence indicating otherwise (several hundred root boles and fallen trunks!) entomological evidence of the widespread Mesolithic and later Neolithic woodlands within the Severn Estuary has been conspicuously absent within the entomological record at Redwick, Gwent (Paddock 2001, 2002; Tetlow in prep) and Gravel Banks, Avon (Tetlow in prep.). The basal peat generally developed later than the submerged forest which it covers, hence the lack of insect species associated with woodland. This balance has, however, been redressed by work at Goldcliff East, which has produced comprehensive and unequivocal woodland assemblages, comparable to those found west of Goldcliff Island by Smith et al (1997, 2000).

Human and animal footprint-tracks and trails (footprint-track sequences) can contribute to many archaeological questions, including human and animal population compositions, hunting practices and husbandry, seasonal occupancy and exploitation, and behaviour in/modification of an environment. A descriptive nomenclature suitable for general archaeological use is outlined, in order to encourage a more systematic approach to the gathering, analysis and interpretation of this kind of important evidence.

For over 200 years after AD 836 the Anglo-Saxon Chronicles and other documentary sources describe Scandinavian activities in the Bristol Channel. Initially this was part of a movement out of the Norwegian homeland by Vikings exploring and raiding through the Scottish Isles and into the Irish Sea. In the South-West Vikings joined in the struggle between the indigenous Britons and the West Saxon kings, targeting royal centres and estates. In the reign of Alfred of Wessex Viking armies were active in the Bristol Channel area as part of a series of attacks which saw his kingdom almost overwhelmed. Attacks on the north Devon and Somerset coast continued in the reign of his son Edward the Elder, who made provision for its defence. No Scandinavian activity is reported on the south coast of the Bristol Channel after 917 until the attack on Padstow in 981. The following twenty years saw war-bands led by Viking kings and jarls active in England. Watchet was one target. The effects were not wholly negative. In the same period the volume of minting from West Country mints, including Watchet, increased and this was probably the product of increased trade in the Irish Sea and beyond, stimulated by the activities of Hiberno-Norse and other scandinavians which also brought silver into the area from outside. In I 068 and 1069, after Harold Godwinson 's defeat and death at Hastings, his sons raised fleets and armies from Hiberno-Norse Dublin to attack the South-West from the Bristol Channel. Their aim appears to have been to raise the South-West against the Norman king William. It failed and they were defeated and driven off, but their activities formed part of a pattern extending back to 836.

Thie paper puts together a proposal for multi disciplinary research strategies and suggests research questions and outcomes that should be met

Construction of a gas pipeline in 1997 between Pucklechurch, east of Bristol, and the gas-fired power station at Seabank, north of Avonmouth provided an opportunity to study an 8 km transect across the alluvial sediments of the Avonmouth Levels. The project produced a more-or-less complete record of the Holocene alluvial sediments along the route of the pipeline; boreholes were commissioned for palaeoenvironmental studies at three locations along the route and palaeoenvironmental data were also collected from excavations of a Prehistoric and Romano-British site close to Easter Compton. The results of the project allow a review of Holocene landscape evolution in the Avonmouth Levels and a re assessment of the evidence for buried soil profiles within the marine alluvium. Collation of data on soil profiles from this and other studies suggests that there are significant problems both with the identification and dating of buried soils within the alluvium.