The Humber Wetland Survey: Vale of York Region

Robert Van de Noort
Humber Wetlands Project
Centre for Wetland Archaeology
University of Hull
Hull
HU6 7RX

Background

The Humber Wetlands Project is a large scale, long term landscape survey project designed to investigate the palaeoenvironmental and archaeological record within the Humber Wetlands area. For practical purposes the area has been divided into a number of sub-regions which are being investigated systematically. To date two reports have been published dealing with Holderness (Van de Noort and Ellis 1995) and the Humberhead levels (Van de Noort and Ellis 1997). Work is proceeding on the survey and publication of the Ancholme and Lower Trent valleys. The Vale of York region, the subject of a separate project design, is thus the fourth area to be investigated.

The study area

The Humber Wetlands survey is primarily concerned with archaeological sites and landscapes where waterlogging has preserved organic remains or where prehistoric landscapes have been covered and protected by marine, estuarine and riverine sedimentation. The study area is thus defined as the area within the Vale of York below a line 10 metres above the Ordnance Datum (AOD). To the south it is bounded by the River Aire and to the north by the city of York. The total area thus defined covers approximately 25,000 hectares.

Geology and landscape history

The history of the landscape of the Vale of York during the Late Devensian and early post-glacial periods is characterised by considerable modification and change. It is in many ways similar to the changes seen in the Humberhead levels which border the area to the south.

During the Dimlington Stadial phase of the Devensian (c. 18,000 to 13,000 BP) an ice sheet covered what is now Holderness and the Lincolnshire Marsh. This blocked the line of drainage through the Humber Gap and impounded the pro-glacial Lake Humber ( which covered much of the Vale of York and the Humberhead levels) shortly before 18,000 BP. In spite of the fact that the ice had retreated northwards by 12,400 BP, Lake Humber remained in existence until about 11,100 BP and only finally disappeared after silting up. The braided rivers which crossed the former lake bed deposited sand and silt levées as they flowed towards the Humber Gap. At about the same time there was a period during which substantial amounts of blown sand was deposited across the lower Trent valley, much of it derived from the reworking of older levées. Radiocarbon dates, together with the evidence of Palaeolithic artefacts from within and beneath the sand indicate that this deposition occurred during the Loch Lomond Stadial phase of the late glacial period, although the evidence of peat deposits dating to the preceding Windermere Interstadial suggests that the process may have begun somewhat earlier.

During the Loch Lomond (Younger Dryas) stage of the Late Glacial, the nick point of the River Humber eroded back through the Humber Gap and triggered a phase of rapid channel incision as the rivers sought to establish a new equilibrium with the low (but rising) sea level. As the sea level rose, incision was replaced by aggradation with the river channels filling with inorganic alluvium and peat. By about 3,500 BP the sea level was roughly equivalent to the modern Ordnance Datum and the deeply cut river channels had been largely infilled. From this time onwards floodwaters tended to spread beyond the channel confines causing overbank alluviation. Much of the alluviation visible today has been attributed to the late Roman period.

Regional vegetational history

Evidence for the Late Glacial vegetational history of the area comes principally from lake sediments found in a hollow of the Escrick moraine near Tadcaster and has been published by Bartley (1962). Evidence from tree and other pollen illustrates the fluctuating climate during the later stages of the Late Glacial period. The earliest post-glacial pollen records indicate a rise in woodland cover, once again with evidence of fluctuations.

Clear evidence for the so-called ‘elm decline’, which occurred across the whole of Britain around 5000 BP, has been found at Burton Salmon where the evidence of other plant pollen seems to indicate an opening up of the woodland canopy with perhaps some clearance for pasture, albeit on a very small scale.

Evidence for the later vegetational history of the vale of York is rather poor, although extensive work on Thorne and Hatfield Moors by Smith (1985) can be used to offset this to some extent.

During the period 4300 - 3600 BP the evidence suggests that the vegetation was dominated by dense mixed oak woodland with pine on the margins of Thorne Moors and on the sandy islands within the developing mires. During the later Bronze Age and into the Iron Age (3600 - 2300 BP) the evidence for clearance increases, with a pastoral economy indicated by the dominant plant species. From the middle Iron Age until the end of the Roman period (2300/2100 to 1650/1450 BP) clearance appears to have progressed rapidly with the evidence of herbs and cereals indicating arable farming. A similar picture is provided by evidence from Crosby Warren on the western edge of the Lincoln Edge. Iron Age clearance of mixed oak woodland was followed by woodland regeneration, with a further phase of clearance during the Roman period. The accumulation of wind blown sand on podzolic soils during the later stages of the Roman period suggests the impoverishment of the soil and severe erosional problems.

The period c. 1550 to 900 BP, which covers the post-Roman period until the time of the Norman conquest, is characterised by a decline in herb pollen and an increase in oak, alder, birch, ash and hazel pollen, indicating woodland regeneration. The following period, beginning around 900 BP, and with an undated termination, sees woodland and scrub clearance and an intensification of agricultural activity.

Drainage history

The courses of the principal rivers which drain the Vale of York, the Aire, Ouse and Derwent, are essentially those which were cut during the late glacial incision phase although in recent times their courses have been embanked and the floodplains substantially drained. It appears that embankment began sometime in the 11th century, although the precise date is far from clear, and the process probably proceeded somewhat irregularly through the medieval period. Drainage gathered pace in the post-medieval period with those associated with Walling Fen and the River Foulness complex being drained after the construction of the Market Weighton Canal in 1772. Today only restricted areas of wetland survive, offering habitats of international importance to migratory birds.

Waterlogged archaeology and buried landscapes in the Vale of York

There are four types of wetland landscape in the Vale of York which have, in general terms, the potential to yield archaeological and palaeoenvironmental data or where landscapes may have been buried by the accretion of sediment.

Areas which were influenced by coastal or estuarine accretion in the past

The lower Aire and Ouse, Walling Fen and the Humber, west of the Humber bridge, appear to have been influenced by estuarine accretion during the Iron Age and Roman periods. In many of the areas affected by this process there is a twofold archaeological potential. Firstly there is the potential associated with the underlying, buried, landscape which may include sites of a dry-land nature and/or sites flooded or affected by a rising water table and preserved in a water-logged condition. Secondly there are sites and finds within accreted sediments. These might be expected to include features associated with wet ground such as trackways, bridges, boats and landing stages.

Major river valley floodplains

This category overlaps to some extent with the previous one, particularly in the case of the lower reaches of river valleys, where estuarine and freshwater accretion may have occurred at different times. In the middle and upper reaches of major rivers, where valley bottom accretion will have been solely of a freshwater nature, the same two-fold potential for preservation will be found as in areas affected by estuarine accretion.

Minor stream floodplains

Minor stream floodplains have a similar potential for preservation as major river valley floodplains, but differ in that they are often better defined and include relict systems which have no modern day water course.

Other areas

In addition to these three major areas of potential there are also other, smaller wetland areas including bogs and fen mires. Although the archaeological potential of these may be limited, their palaeoenvironmental potential is considerable.

Previous archaeological work in the Vale of York

The first stage of the Humber Wetlands Survey involved a comprehensive assessment of the existing state of knowledge of the areas to be investigated (Van de Noort and Davies 1993). This used data from county Sites and Monuments Records (SMR) and demonstrated the extremely high archaeological potential of the area. It also suggested that the known sites represented only a small part of the archaeological potential of the area as a whole. The Vale of York suffered from many of the same problems as the other sub-regions, while at the same time offering a similar degree of potential. Specific problems included a poor representation of Palaeolithic and Mesolithic sites and finds dominated by stray finds of individual objects. Later prehistory and the Roman period are better represented, although problems of patchy research remain.

The post-Roman period appears to be seriously under-represented and the medieval mainly by standing buildings and pottery scatters on the edges of the floodplains.

Threats to the archaeology of the Vale of York

In the initial survey of the area covered by the Humber Wetlands Survey four specific threats to the survival of the archaeology were identified:

• Land drainage and changes in agricultural land use
• Coastal and estuarine erosion
• Mineral extraction
• Industrial development

In the Vale of York the first and second threats are of particular concern and merit some further description.

Land drainage and changes in agricultural land use

While the continued drainage of agricultural land has affected all wetlands in the Vale of York, those in the River Foulness / Walling Fen complex have been particularly severely damaged, with drainage ongoing since the construction of the Market Weighton canal in the late 18th century.

The river floodplains of the Ouse, Derwent and Aire are somewhat better protected, although this is patchy. In general the impact of land drainage and changes in land use vary across the region, with areas such as the Derwent Ings relatively well protected, but with advanced desiccation affecting other areas.

Estuarine erosion

Estuarine erosion has long been recognised as the main threat to the archaeologically rich Humber foreshore, which includes North Ferriby and Brough where boats and shoreline features dating to the later Bronze and Iron Ages have been found. Continued monitoring at North Ferriby has demonstrated a lateral erosion rate of 0.5m per year, varying with the state of the main channel of the Humber. Concern over the general erosion of the foreshore has led the Environment Agency to design a flood management plan for the estuary as a whole. While the plans are generally linked to existing defences and embankments, it is expected that there will be impacts on the archaeologically sensitive areas, particularly through the use of heavy machinery on and adjacent to the various sites

Aims and objectives

The overall aims and objectives of the Humber Wetland Survey have been set out by Van de Noort and Davies (1993). The programme developed for the Vale of York draws extensively on these and additionally draws attention to the particular requirements of the area in its own right. These include

• The diversity of landscape types in the Humber Wetlands area as a whole means that there is a strong likelihood of a broad chronological range of waterlogged sites.
• The same diversity of site types and dates means that there is an increased chance of unusual and poorly represented sites and finds being preserved. The prehistoric boats from the Humber foreshore are one example of this.
• The Vale of York remains a relatively poorly known area, in comparison to other areas of the Humber Wetlands. The example of the previous surveys suggests that this situation will change dramatically after the proposed programme of research is carried out.
• The use of a survey methodology has the advantage of relating sites and finds through their landscape context. The palaeoenvironmental element further enhances this contextual approach by locating the archaeological data within its environmental matrix.
• The surviving wetland sites and areas within the Vale of York are threatened by a similar range of modern farming and land use practices as those seen elsewhere in England (including the Somerset Levels, the Fens and north-west England).

In Exploring our Past (English Heritage 1991) a number of priority research themes were identified. Amongst these were ‘processes of change’ (1991:35-37). A number of the specific themes within this broad category are of particular relevance to the Vale of York and will be addressed during the programme of survey and the subsequent analysis:

• Hunter-gatherers into farmers; building on the large number of late-Mesolithic and early Neolithic sites located during the first three programmes of survey, this will be the subject of further research within the Vale of York.
• Briton into Roman; the Vale of York offers a particularly suitable region for the investigation of this topic with a number of Late Iron Age and early Roman sites (notably industrial sites) known from Holme-on-Spalding Moor and elsewhere.
• The early medieval period; a period very poorly known outside the city of York itself, but with the results from the city offering the opportunity for comparative study.
• Landscape form and change during the Holocene; this is a central element in the overall project and one which will be further pursued in the Vale of York

The Vale of York survey will continue ongoing programmes of work concerned with the impact of drainage and changes in hydrological regimes on archaeological and palaeoenvironmental sites, as outlined by Van de Noort and Davies (1993), with the aim of providing information of use in planning and management matters. Liaison with the English Heritage Ancient Monuments Laboratory (AML) will continue over a number of issues including the stability of waterlogged remains in the ground and dendrochronology.

Other collaborative work will include a programme of palaeogeographic mapping linked to the NERC funded Land Ocean Interaction Study and carried out as a joint project between Hull University and English Heritage. In addition two research studentships have been awarded by the University of Hull to the project and a third by Sheffield University. These are focused on the effects of sea level change on the wetlands in the upper Humber estuary, the early agricultural exploitation of the area and the late Mesolithic to early Neolithic transition in Holderness. Further research projects will be brought forward for consideration during the course of the Survey.

Methodology

The overall methodology applied by the Humber Wetlands Survey team has been set out in the original project design and examples presented in the two initial reports (Van de Noort and Davies 1993, Van de Noort and Ellis 1995, 1997). In general terms the strategy to be applied in the Vale of York will follow the established pattern, with adjustments made in the light of recent experience.

Prospecting methods to be used in the Vale of York will be based on three complementary techniques.

Fieldwalking

Fieldwalking will be conducted at thirty metre intervals on the wetland margins and the wetlands themselves. Attention will be paid to areas where desiccation has reduced the ground surface and exposed the underlying landscape. Systematic survey will be undertaken in approximately thirteen ‘mapviews’ each measuring four by five kilometres.

Transect coring

Transect coring will be used to assess the smaller river valley floodplains and palaeochannels. The coring exercise will be undertaken within the same ‘mapviews’ selected for fieldwalking, allowing the close integration of the different types of data.

Dike survey

Dike survey, although less effective than hitherto as a result of environmentally friendly dike cleaning practices, will be used wherever possible.

Artefact recovery

Artefacts found during the survey will be processed concurrently with the fieldwork to allow the rapid feedback of information to the field team. The collection policy is designed to produce a representative sample of material which will enable the identification, dating and characterisation of the site in question. Late and post-medieval pottery will not be retrieved if found in low densities as this is considered to represent the results of manuring and to have little relevance to the study of wetlands and their exploitation.

The initial assessment of archaeological sites

An assessment of up to ten sites will be undertaken with a view to establishing their date, extent, character and state of preservation. The sites will be selected from known sites and landscapes and from the sites discovered during the fieldwalking programme. Assessment will involve fieldwalking, coring and, where appropriate, the use of test pits. Samples will be taken to assess the preservation of pollen and wood and for other types of analysis as appropriate. Selective use will also be made of geophysical survey, in association with the Archaeometry branch of the Ancient Monuments Laboratory.

Two known sites have been pre-selected for assessment:

• The North Ferriby foreshore
• The Melton foreshore

Both of these sites have produced wooden artefacts (including, in the case of Ferriby, boats) and indications that wooden structures of later prehistoric date survive.

Palaeoenvironmental programme

A programme of palaeoenvironmental fieldwork, focusing on buried landscapes, will be integrated with the field survey and site assessments. This is designed to improve and expand upon existing data and to assist in the production of palaeogeographic maps of the region. Strategies will include the mapping and characterisation of buried land surfaces, the laboratory analysis of cores and the location of deposits of ages or types not previously studied from a palaeoenvironmental perspective. This programme will have three main components:

• The location, mapping and characterisation of the former drainage networks of the Ouse, Derwent, Aire and lower Wharfe valleys.
• The detailed mapping of the surviving prehistoric landscape along the rivers Humber and Ouse.
• The palaeoenvironmental assessment of waterlogged deposits dating from the historic period.

Publication

The format for the publication of the results of the sub-regional surveys has been established by the work in Holderness and the Humberhead levels (Van de Noort and Ellis 1995, 1997) and it is expected that the Vale of York survey will yield a similar style of volume. In addition a third brief annual report will be produced outlining the work undertaken and the results obtained.

Bibliography

English Heritage 1991 Exploring our past. Strategies for the archaeology of England. English Heritage.
Bartley, D.D. 1962 The stratigraphy and pollen analysis of lake deposits near Tadcaster, Yorkshire. New Phytologist 61: 277-287.
Smith, B.M. 1985 A palaeoecological study of raised mires in the Ancholme and lower Trent valleys. Unpublished PhD thesis. University of Wales
Van de Noort, R. and Davies, P. 1993 Wetland Heritage; an archaeological assessment of the Humber wetlands. Humber Wetlands Project.
Van de Noort, R. and Ellis, S. 1995 Wetland heritage of Holderness; an archaeological survey. Humber Wetlands Project
Van de Noort, R. and Ellis, S. 1997 Wetland heritage of the Humberhead levels; an archaeological survey. Humber Wetlands Project