Humber head levels

The Wetland Heritage of Humber Head Levels

The Humberhead Levels comprise three different types of wetland, undoubtedly the best known of which are the raised mires of Thorne and Hatfield Moors, which together measure approximately 3300 ha. Despite their status as Sites of Special Scientific Interest (SSSI), and the recent declaration of Thorne Moors as a National Nature Reserve (NNR), they suffer from severe desiccation as a result of drainage, while the ongoing peat extraction over much of the Moors is continuing to damage the palaeoenvironmental resource provided by the peat and the archaeological remains contained within it.

The second type of wetland lies to the north and east of Thorne Moors and is known as Marshland. This comprises approximately 6000 ha and has been created and dominated by estuarine accretion in the past. The accumulation of sediment in this area continued well into the present century, a process encouraged by landowners to improve the quality of their land and commonly referred to as 'warping'. Despite the burial of prehistoric and historic landscapes beneath such sediments, the intensive character of agriculture in the area is likely to threaten the potential archaeological resource in the near future.

The third type of wetland comprises the river floodplains of the Aire, Went, Don, Torne and Idle, and their tributaries such as the Hampole Beck. Although their individual histories and impact on the landscape vary greatly, due to the rise in sea-level during the last 10,000 years, all the rivers experienced an initial phase of incision when sea-level was more than 20 m below its present position, followed by a phase of aggradation, with the accumulation of alluvium and peat, and overbank flooding resulting in large-scale peat development in the area south and east of Thorne and Hatfield Moors. River management, starting with the ambitious drainage schemes of Cornelius Vermuyden in the 1620s and continuing up to the present time, combined with field drainage and modern farming techniques, has already destroyed much of the archaeological resource in this type of wetland and continues to threaten much of what remains.

Fieldwork in the Humberhead Levels has involved three main elements - the archaeological field survey itself, palaeoenvironmental work and the assessment of the state of preservation of a limited number of archaeological sites.

Rossington Roman Bridge Site

roman postThe same Roman road which was assessed at Scaftworth travels across the Idle to Bawtry and turns northwards towards Danum. It crosses the River Torne at Rossington, at a site already well known for the existence of more than 10 Roman pottery kilns. A site survival assessment was undertaken in an area where in the past groups of squared posts had been seen, evidently the foundations of a Roman bridge. The site assessment, undertaken in September 1995, retraced at least one of these groups of squared posts. Their preservation had been limited to the area below the lowest water-level, the gravel surrounding the posts not providing any protection from the seasonal fluctuations in water-level. Although it is difficult to determine the rate of desiccation, we must assume that as much as 1.5 m of the posts have deteriorated through desiccation since the 1950s, when the area was studied by Doncaster Museum.

This site assessment is of particular importance because of the proximity of the Roman kilns site, a Scheduled Monument which is known to have contained extensive well preserved waterlogged remains. The site survival assessment not only indicated the effects of de-watering on the area, but during the assessment a two-week-long fire in the peat of the scheduled area was seen to destroy much of the soil and any palaeoenvironmental and archaeological organic material contained within it.

Scaftworth Roman Road Site

reconstruction of road siteAt Scaftworth, an assessment was undertaken on the Roman road which crosses the River Idle from Nottinghamshire into South Yorkshire. It has been assumed that this road (Iter V of the Antonine Initiary) from Lindum to Eburacum via Danum was built in the second half of the first century AD, possibly in or around AD 70 during the early stages of the conquest of the Brigantes, who up to that date had been allied to the Romans since AD 43. The route of the road has been known for some time, and in 1983 the crossing over the Idle floodplain was discovered after drainage works. The site was initially excavated in 1983 by the University of Sheffield, and subsequently assessed in 1991. Finally, in 1995 a full-scale assessment was carried out as part of the Humber Wetlands Survey.

excavation of road siteA trench 4 m wide and 9 m long was excavated. This clearly showed that not one but two roads existed here, both crossing the Idle floodplain but with slightly different alignments. The first road was constructed of wood and turves, and the reconstruction drawing on the cover of this report is an artist's impression of the actual building of the road. It consisted of three parallel rails, over which smaller timbers were placed. All timbers came from alder and willow/poplar trees, which were readily available on the floodplain. The timbers had been felled with axes and a few pegs were used to keep the construction in place. On top of the timber 'raft' a concave surface was created by turves, which came from drier land nearby. Analysis of the macrofossil remains beneath this road suggests that the area was very wet, which is reflected in the drawing. Although it remains to be proven, we assume that the road may have been the first to be built here and datable to the first century AD. If this was the case, the construction would have been carried out by legionary soldiers.

At some later date, possibly in the third century AD, a second road replaced the timber and turf construction. It consisted of alignments of oak pegs on either side of the road, with a gravel body in between, possibly raising the surface to create a causeway over the floodplain. Large oak piles appear to have been used to keep the gravel in place. A thin layer of peat within the gravel bed suggests that the road was repaired or made higher on at least one occasion after its construction. In the post-Roman period, the river re-worked the gravel and spread it out, resulting in an 18 m wide gravel ridge. A leather sole of post-Roman date suggests that this ridge may have been used well into the Middle Ages as a route to cross the Idle.

In terms of survival of waterlogged remains, the work showed that the effects of desiccation and ploughing had resulted in ongoing deterioration of the timbers and of the peat surrounding the road. The oak timbers of the later road in particular were badly desiccated and riddled with roots, which could not penetrate the gravel metalling. The horizontal timbers of the earlier road were better preserved, but the top of the road had deteriorated significantly more than the bottom, a sure sign that the effects of drainage now threaten the whole road. The peat beneath the earlier road was still well preserved, but the peat which had built up around and above the later road (of particular value as peat of the last two millennia is extremely rare) was desiccating rapidly, leaving the gravel ridge clearly visible as an upstanding feature.

Back to top