Collapse or not? Effects of drying on heavily decayed archaeological wood in soil

The physical response of heavily decayed archaeological wood to water evaporation in soil was studied in a model experiment. Ten wood samples from a kitchen midden at Kangeq in Western Greenland and four wood samples from
the Danish bog site Nydam were used in the experiment. Microscopic analysis showed the secondary cell wall was totally decayed in all samples. The samples from Kangeq were decayed by wood degrading fungi and the samples from Nydam were decayed by bacteria. Following characterisation, the samples were buried in moist soil that was allowed to slowly dry out in the laboratory by
evaporation at room temperature with sensors measuring changes in soil water potential and soil moisture. The mass, radial shrinkage and cell wall collapse of the individual samples were determined after 26, 48, 56, 62 and 82 days in the soil. During this period, the water content of the soil decreased from 48 to 22% vol and the water potential dropped from 0 to -65 kPa, corresponding to a decrease in relative humidity (RH) within the soil from 100 to 99.95% RH. The archaeological wood from Nydam showed a high radial shrinkage and cell collapse, while samples from Kangeq only showed moderate radial shrinkage and no cell collapse with decreasing water content of the soil. The absence of cell collapse in the wood samples from Kangeq is surprising considering the degree of degradation of the secondary cell wall compared to the prevalent
collapse theory of degraded archaeological wood.