Carbon-isotope analysis of fossil wood and dispersed organic matter from the terrestrial Wealden facies of Hautrage (Mons Basin, Belgium)

Abstract

Pedogenic nodules, calcitic shells, dispersed organic carbon and bulk rocks affected by low diagenesis are judged to be reliable tools for carbon-isotope chemostratigraphy. Up to now no systematic study has been carried out to check the reliability of fossil wood material for carbon-isotope chemostratigraphy in terrestrial settings. The 235-m thick Wealden facies sediments of the Danube-Bouchon quarry at Hautrage (Hautrage Clays Formation, Mons Basin, Belgium) encompass dark to grey clays and sands, rich in organic matter and coalified–charcoalified fragments of fossil wood, deposited in an alluvial plain during middle Barremian to earliest Aptian. We measured and compared the carbon-isotope ratios of 110 levels of the stratigraphic succession for dispersed organic carbon ( δ 13C DOC), and fossil wood ( δ 13C WOOD) collected in the same geological level. In the whole succession, the averages of δ 13C DOC and δ 13C WOOD have a significant difference of about 0.9‰. The δ 13C WOOD is usually heavier than the δ 13C DOC, which is consistent with measurements on different constituents of modern trees (branches versus leaves and tissues). In one single stratigraphic level, the variability of δ 13C WOOD is much higher (up to 7.3‰) than that of δ 13C DOC (0.4‰). Four main causes may explain these results. Firstly, the δ 13C DOC averages the isotopic signal of different compounds and tissues (such as leaves, seeds, cuticles) as they become mixed and dispersed. The δ 13C WOOD reflects the carbon-isotope ratio of a small part only of one single tree, which is a complex system with δ 13C variations over a range of 4‰. Secondly, there is strong δ 13C variability between different species of plants. In Hautrage, several gymnosperm genera were collected (including Podocarpoxylon, Taxodioxylon, Brachyoxylon, and Thujoxylon) and numerous fern taxa. Thirdly, coalification and charcoalification can affect the δ 13C of the DOC and the various kinds of woods in a different way. Fourthly, the wood fragments may be reworked several times from more ancient geological levels, especially in the alluvial plain environment of Hautrage where thick levels of fluviatile coarse sand deposits are observed. However, in the whole succession, both δ 13C WOOD and δ 13C DOC curves show similar trends. This suggests that carbon-isotope curves on fossil wood can be matched to carbon-isotope curves on DOC. For the whole trend fossil wood is a relatively good chemostratigraphic tool when sufficient samples are measured in the succession. Whenever possible both control of the wood taxa, and estimation of degree of (char)coalification are however recommended. In one single level, charcoal has more stable δ 13C WOOD values than δ 13C WOOD of coalified fragments. The δ 13C positive trend can be due to several causes, including global pCO 2 variations and/or regional changes and/or local environmental conditions in the alluvial plain. If global, the dating of the Hautrage succession may be refined to the late Early Barremian–early Late Barremian by matching palynological and carbon-isotope chemostratigraphical data.