Olympics Site 25, New Aquatics Centre, Carpenters Road (OL-00105)

Clients: Capita Symonds

Author: Jane Corcoran

Site supervisor: Isca Howell

Ten evaluation trenches were excavated across the site during February and March 2005. Previous geoarchaeological deposit modelling, undertaken as part of the desk based assessment, indicated that the site lay close to the deepest part of the ancient valley floor. However, a projection of the London to Colchester Roman road alignment also suggested that the road might cross the southern part of the site. The evaluation was intended to look for evidence of river channels, landscape change and wetland archaeology and establish whether any evidence for the road or its river crossings existed on the site.

The joint MoLAS-PCA excavating team included a geoarchaeologist. This was essential as, in river valleys, thick alluvial deposits exist that represent a range of wetland and dryland environments and archaeology is likely to lie within this natural deposit sequence, as well as being cut into it and resting on its surface. In order to interpret the archaeology an understanding of the environments represented by the surrounding deposits is needed. From the geoarchaeological point of view, it is much more reliable to interpret a deposit from exposed sections than to be given samples from it, taken by the excavators, for examination and interpretation off-site. In fact, it is often found that sampling is not actually needed, as sufficient information can be obtained by on-site examination of deposits to interpret them. Instead, targeted sampling can be undertaken to address specific questions and clarify uncertainties in the deposit sequence, as well as enabling the best places in terms of their likely environmental potential and significance to be sampled.

A sequence of alluvial deposits was found that recorded the lateral migration of a tributary channel of the Lea during the prehistoric period, which has been dated by five radiocarbon samples. The base of the alluvial sequence was not seen, but previous geotechnical boreholes suggest it lies at around 1m below the base of the trenches and the lowest levels of the alluvium will be examined as a later stage of work. The earliest deposits observed were tufa-rich sands accumulated in clear flowing water, when the climate was warmer than today, during the Early Neolithic and before. Butchered bone was found in a peaty land surface of Neolithic date and worked wood, dating to the Bronze Age, had been washed-up as driftwood at the margins of the later river.

The Roman road was not found. However, further radiocarbon dating of the upper layers of the channel deposits is being carried out and this should establish whether the site was still a wetland area associated with the river in Roman times, or whether the clays that form the upper part of the alluvial sequence across the entire site were already being deposited. The characteristics of the clay are indicative of seasonally flooded grassland, suitable for grazing, which probably existed on the site throughout the historic period. A buried soil and turf line at the top of the alluvial clay indicates that river straightening and constraining reduced the episodes of prolonged seasonal flooding, allowing a drier land surface to develop prior to the 19th-20th century industrial development of the site. Two metres of made ground, heavily contaminated with the chemical products of this industrial activity, sealed the buried soil.

Environmental samples, comprising blocks of undisturbed sediments (monoliths) and adjacent bulk samples, were taken from marshy backwater areas where peaty sediments had accumulated behind successive sand banks. These samples are currently being assessed and the results should guide the sampling strategies on future sites in the Lower Lea, in particular those that will be excavated as part of the Olympics scheme.

This site report is extracted from MoLAS 2005: annual review