Project Details
Projekt Print View

Geochemical analyses of sediments to reconstruct landscape changes and settlement dynamics in the western Nile Delta, Egypt

Subject Area Physical Geography
Term from 2018 to 2022
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 392434407
 
Final Report Year 2024

Final Report Abstract

Archaeological surveys and excavations in the western Nile Delta have shown that, contrary to earlier assumptions, this region has been inhabited since the 4th millennium BC at the latest. Research at the archaeologically significant site of Buto (Tell el-Fara'in) and its surroundings has revealed settlement phases in the pre-dynastic and early dynastic periods, the Old Kingdom and from the Third Intermediate Period up to the Islamic period, while settlement traces from the Middle and New Kingdoms are largely missing. The aim of the project was to use a geoarchaeological approach to investigate whether changes in environmental conditions and the fluvial network influenced the attractiveness of this area for settlement and use. Holocene Nile sediments are excellent geoarchives that allow conclusions to be drawn about environmental changes. This requires sediment analyses and, above all, reliable age determinations (e.g. 14C, OSL). The latter are only feasible to a limited extent or not at all in Egypt. However, the export of samples is prohibited by restrictions imposed by the Egyptian authorities. Therefore, a further aim of the project was to develop and test a research approach primarily based on geochemical analyses using a portable XRF device. Using complex methods of data handling and processing as well as machine learning techniques, environmental changes are to be recognized and classified over time. On the one hand, the element compositions of Nile sediments provide information about the respective depositional environments. On the other hand, the geochemical data contain anthropogenic signals, which vary depending on different colonization phases. Due to the large amount of data, machine learning methods, such as artificial neural networks, are used to assign sediments from corings to specific cultural periods based on their specific geochemical characteristics. Geochemical analyses of settlement remains, which are reliably dated by cooperation partners from the field of archaeology, serve as training data. In the course of the project, the database already established during the preliminary investigations was expanded through drilling and analyses during three field and laboratory campaigns. In total, analysis results are available for more than 2660 samples from 65 boreholes from the region around Buto. On this basis, together with the sediment records, supplementary geoelectrical measurements and remote sensing methods, it was possible to partially reconstruct the former river and canal system. The use of various machine learning methods also allowed the rough chronological classification of channel fills and Nile sediments. In order to achieve a higher temporal resolution, the training data set should be extended by geochemical analyses of more than the three cultural epochs to date. Although this depends on the results of the archaeological activities in the study area.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung