Deciduous oak/chestnut: Differential shrinkage of wood during charcoalification? Preliminary experimental results and implications for wood diameter study in anthracology

Recent methodological developments in dendro-anthracology offer many new techniques to address the historical practices associated with the exploitation of wood as a fuel and to better understand the reciprocal interests of forest landscapes and societies. The study of wood diameter in dendro-anthracology involves three fundamental stages: i) to measure the position of charcoal pieces in relation to the heart of the wood, known as charcoal-pith distance ii) to correct the data, taking into account the error associated with using measurement as a tool and wood shrinkage, iii) to interpret the data in terms of original diameters exploited. Each of these steps has been, and continues to be, the basis of numerous research projects, notably by using experimental approaches in order to perfect the pith estimation tools (used to measure the distance between the charcoal fragment and the centre of the stem, refine experimental techniques and improve methodologies. This article is focused on estimating suitable correction factors related to wood shrinkage. Although this phenomenonis very well understood for a great number of species as regards the drying process, it remains more difficult to quantify a precise correction factor for shrinkage during the wood burning process. The experimental approach presented in this study involves a comparative analysis of wood shrinkage in oak and chestnut in charring conditions close to those observed in a charcoal-making context (charring in a low-oxygen environment, at temperatures between 400 degrees C and 600 degrees C). The charring of wood sections in a muffle furnace clearly shows an increase in the difference in radial shrinkage between dried and charcoalified wood. During drying, shrinkage in oak is only 2% greater than in chestnut. By contrast, the difference in shrinkage between the two taxa is 7% after charring. In addition, these experiments highlight the significant influences of temperature and of the presence of heartwood on the extent of wood shrinkage that occurs when wood is burned. The results of this study provide a better understanding of the phenomenon of shrinkage and also enable correction factors for these two taxa, more suitable for application in an archaeological context, to be proposed. (C) 2017 Elsevier Ltd and INQUA. All rights reserved.