STEROLS AND BILE ACIDS IN URBAN AND RURAL SOILS AS FAECAL MARKERS OF LAND-USE SINCE THE BRONZE AGE
Résumé
For ancient societies, the control of agriculture and animal domestication were extremely important to their development. After having domesticated different animals as livestock, ancient societies developed techniques such as manuring to improve culture yields. There are numerous evidences in archaeological records of animal breeding (bones, coprolithes, spherolites). Reversely, despite few studies dedicated to buildings and caves, the only evidence for manuring and breeding out of archaeological context, were acquired on recent soils (Bull et al., 1998; Birk et al., 2011). The aim of this study is to test the potential of faecal biomarkers preserved in soils to spatially resolve (at the field-scale) the strategies of land use (buildings, breeding and manuring) in ancient times. For this, faecal biomarkers (sterols and bile acids) of animals (Humans, cow, horse, pig and sheep), sediments from a septic tank (beginning of 19th century) and soils of distinct land use (pasture and forest) were characterized. Then, archaeogical samples were analysed: (i) agricultural soils, middens and settlement layers dated back to the Bronze Age in Lake Bourget shore, (ii) settlement layers recovered during excavations in a medieval building (15th century, Orléans). Sterols and bile acids were isolated from archaeogical samples following Zocatelli et al. (2012). Neutral fraction was further separated by flash chromatography. The alcohol fraction was silylated and analysed by GC-MS with a Polaris TRACE-GCQ. From the twenty sterols and bile acids identified, the following faecal biomarkers were found: coprostanol (Cp), methyl-coprostanol (mCp), ethyl-coprostanol (eCp), lithocholic acid (LC), deoxycholic acid (DOC), cholic acid (C), chenodeoxycholic acid (CDOC), hyodeoxycholic acid (HDOC), ursodeoxycholic acid (UDOC) and 3α-hydroxy-12-oxo-5β-cholanoic acid (oxoDOC). To identify possible sources of faecal compounds, bile acid imprints and sterol ratios were employed: 1) (Cp+epi-Cp)/(eCp+epi-eCp); 2) (Cp+epi-Cp)/(Cp+epi-Cp+5α-cholestanol); 3) (Cp+epi-Cp)/cholesterol. Sterol content of animal faeces clearly distinguished two groups: omnivores (Humans and pig) and herbivores (cow, horse and sheep). Then, Human and pig imprints were distinguished from bile acids. Faecal biomarkers were detected in all archaeogical samples (i and ii) and allowed the distinction between omnivores, herbivores and mixed faecal biomarker imprints in the sample set. Samples from middens of Lake Bourget catchment contained a mixture of Human and breeding faecal sterols. Bile acids corroborated these results. Furthermore, faecal biomarkers detected Le Bourget catchment soils formerly cultivated for millet, evidenced past manuring practices (mCp, eCp and DOC). High contents of coprostanols (Cp and eCp) and bile acids (LC, DOC and C) in settlement layers from Lake Le Bourget shore evidenced the co-occurrence of animals in the same area. Analysis of faecal materials from a medieval building in Orléans revealed only Human occupation (Cp, LC, DOC, CDOC, C and oxoDOC). Finally, this study showed that faecal biomarkers are efficient to provide information on the use of space in areas inhabited by humans. From a more general point of view, these results illustrateed that faecal biomarkers could reveal how Human societies shaped their environment through times.
Origine : Fichiers produits par l'(les) auteur(s)
