Anthracological research in the Fertile Crescent

Dr. Katleen Deckers

The project aims to understand vegetation changes in the Near East from charcoal remains from archaeological sites in Syria and Turkey. In large areas of the Fertile Crescent, pollen preservation is bad due to dry and alkaline conditions and therefore palynological research often impossible. Where pollen were still preserved and studied, large problems exist with the chronology of vegetation changes due to the “reservoir effect”. Therefore, charcoal remains from archaeological sites within this area can deliver important new information on the palaeo-environment and use of wood.

On one hand, the project focuses on the transition from Pleistocene to Holocene, based on charcoals from the archaeological sites of Baaz and Kaus Kozah in Western Syria (Figure 1). Up to now, in total 11447 charcoal fragments have been identified from Baaz. The anthracological research indiates that during the Younger Dryas the site was located within the almond-pistachio steppe, probably relatively far away from the oak-Rosaceae park woodland and the dense stands of wild cereals. The find of charcoal from Populus (poplar)/Salix (willow) indicates the presence of a permanent water source in the neighbourhood of the site, which must have paled an important role for the site location. Preliminary analysis of samples from Kaus Kozah also indicates large proportions of almond wood.

On the other hand, the project concentrates on gaining insight into the palaeo-vegetation and human use of it from Bronze through Iron Age. For this, charcoals have been investigated from Tell Mozan and Tell Leilan in northeastern Syria, Emar on the Middle Euphrates in Syria, Tell Jerablus on the Euphrates in Syria near the Turkish border, Qatna in western Syria, and Tell Atchana and Kinet Höyük in southern Turkey (Figure 1).

Up to now, ca. 10000 fragments have been identified from the Tell Mozan settlement excavations. The dominating presence of deciduous oak charcoal in most of the samples, the presence of charcoal remains from small oak branches as well as the find of acorns suggests that the region was vegetated by an oak park woodland. Today, Tell Mozan is however located within a tree-less agrarian steppe (Figure 2). During the Bronze Age, the deciduous oak must have had a more southwards distribution than today (Deckers and Riehl 2007).

From Tell Leilan, ca. 2400 charcoal fragments have been identified. Populus (poplar)/Salix (willow) (including small twigs) dominates within the samples and may have been present locally along the Wadi Jarrah. Fraxinus (ash) and Ulmus (elm), are also often present, and may have belonged also to the Jarrah riverine vegetation. Besides riverine vegetation, deciduous oak is dominating, which supports the results from Mozan of a more southwards distribution of deciduous oak in the past. The samples contain however more Pistacia (pistachio) and Amygdalus (almond), suggesting the pistachio-almond woodland steppe was closer to the site than at Mozan.

The more southwards distribution of deciduous oak also has been found at Tell Jerablus on the Euphrates, where a component of deciduous oak fragments was found amongst mostly riverine taxa. From this site, up to now a 6000 charcoal pieces have been identified, mainly from hand-picked samples, but work on flotation samples is in progress.

Further south, the rich Bronze Age Euphrates vegetation has been documented at Emar. The analysis of more than 24000 charcoal fragments from this site suggest that during the Bronze Age, the riverine gallery forest was more extensive than today and consisted of a greater variety of taxa, including Populus /Salix, Tamarix (tamarisk), Alnus (alder), Fraxinus, Platanus (plane) and Ulmus. The discovery of Olea (olive) in Early Bronze Age layers, Vitis (vine) within Late Bronze Age strata and Ficus (fig) wood in Early and Late Bronze Age layers might indicate their cultivation. There is little evidence for the potential terebinth-almond woodland steppe away from the Euphrates, which suggests that it was probably degraded. Coniferous wood, like Cupressus (cypress), Cedrus (cedar) and Pinus brutia/halepensis (Calabrian/Aleppo pine) were probably imported (Deckers 2005).

The analysis of charcoals from sites located in the Mediterranean zone is in progress.

At Qatna a total of 4400 charcoal fragments from Early Bronze to Iron Age contexts have been identified so far, with a majority from Late Bronze Age strata. Overall, 15 taxa have been identified. Deciduous oak is present in all strata and probably represents the oak-Rosaceae park woodland in which Qatna was probably located. While no single taxon dominates in the Early and Middle Bronze Age, more than 70% of the identified fragments from the Late Bronze Age and Iron Age is cedar. This change in pattern may be a sympthom of deforestation, alternatively may reflect different contexts.

From Atchana, ca. 1000 charcoal fragments from almost 70 samples have been identified. Oak dominates and is present in deciduous and (less)evergreen varieties. Pinus brutia/halepensis is also one of the main taxa, besides Olea. Some Euxian/Medio-European taxa were found as well: like e.g. Taxus (yew), Carpinus and possibly Fagus (beech).


Initial phases of the project have been supported by a DFG-post-doctoral fellowship in the Graduate College “Anatolia and its Neighbours” . Additionally research has been supported by a project grant from the Strukturfonds of the University of Tübingen (project direction Dr. Simone Riehl). The Belgische Stichting Roeping supports the project. Many thanks are due to the excavation directors which allowed investigating the charcoals from their sites.