The Experimental Earthwork at Wareham, Dorset after 33 Years: 3. Interaction of Soil Organisms with Buried Materials

Abstract

The Wareham Experimental Earthwork was constructed in 1963 in an area of heathland in the south of England with acidic sandy soils to investigate the processes that occurred early in the establishment of the archaeological record. Amongst its objectives was monitoring the changes to various archaeological materials that were buried in the earthwork. In this paper we present data on the interaction of soil micro-organisms with linen, flax, goatskin and hemp buried in 1963 and recovered during the 1996 excavation of the earthwork. These materials were originally buried on the old land surface beneath a stack of freshly cut turves (the turf environment), where the total C and N contents were relatively high, and higher up in the earthwork (the sand environment). In the 33 years since the earthwork was constructed all the visible remains of the linen, flax and hemp had been lost except where preservation had been aided by the presence of copper alloy or steel discs. The size of the soil microbial community (microbial biomass) and microbial activity (respiration rate) were greater in the turf environment than in the sand environment, but all the values were very small compared with typical ranges for soils. There was some evidence, in the form of increased respiratory responses to added glucose and greater respiration rates, that a legacy from the flax and the hemp persisted after they ceased to be visible. However, the results from adenosine triphosphate (ATP) analysis, which provides a measure of physiologically active micro-organisms, and from substrate utilization profiles (which provide an indication of the metabolic capability of the microbial community) did not support this. A short-term (7 months) laboratory experiment using samples of the original materials from the experimental archive and soil samples from the earthwork was conducted to attempt to simulate the conditions at the outset of the experiment. The rate of decomposition of the materials during this experiment was initially rapid and after 7 months between about 10 and 20% of the carbon in the buried materials had been lost. The substrate utilization profiles of the microbial communities associated with the decomposing materials could be related to their chemical composition, with the communities associated with plant-derived (carbohydrate-rich) materials giving large responses to carbohydrates, and the goatskin (mostly proteinaceous tissues) giving large responses to amino acids (components of proteins). Solid-state nuclear magnetic resonance (NMR) spectroscopy was used to investigate changes in the distribution of carbon between differentfunctional groups indicative of some of the main types of biochemicals in the materials during the short term decomposition experiment. Only small changes were observed in the NMR spectra of the plant-derived materials. Consistent with the substrate utilization profiles, there was evidence of protein depolymerization to amino acids in the NMR spectra for the decomposing goatskin. When the materials were incubated in a reference soil that contained more N and a larger community of active micro-organisms than either of the soil types from the earthwork, the short term decomposition rate of all the materials was more rapid than in the earthwork soils, with the increase in decomposition rate being more marked for the plant-derived materials. It is hypothesized that this observation is due to a lack of available N restricting decomposition of the plant-derived materials in the earthwork soils.