Geomagnetic field variations and low success rate of archaeointensity determination experiments for Iron Age sites in Bulgaria

Abstract The geomagnetic field variations during the Iron Age (last 1200 years BC) are poorly constrained for Bulgarian territory due to the scarcity of archaeological sites with a well-established chronology. Additionally, the collected Iron Age materials frequently fail to provide reliable archaeointensity data. In an effort to expand the Bulgarian archaeomagnetic database for the last millennia BC and to find an explanation for the low success rate of Thellier experiment, 28 baked clay structures from 10 archaeological sites were archaeomagnetically studied. Magnetically soft minerals dominate in all collections despite the variety of the sampled structures (hearths, domestic ovens, ovens, pottery kilns, burnt dwelling remains, mud-bricks), which suggests differences in the ancient firing conditions. The main carriers are magnetite/titanomagnetite and/or maghemite/titanomaghemite. Verwey transition is well expressed in most samples. Usually more than one magnetic phase exists and noticeable mineralogical transformations occur during laboratory heating. Regardless of all the pre-selection criteria applied, the total success rate of intensity determination is quite low (21%), which seems to be typical for Bulgarian Iron Age materials. At this epoch that starts with a significant decline in the quality of material culture, combustion structures are usually poorly preserved, short-term used (frequently only once) and heated to temperatures around and below 600 °C. A clear relationship is observed between median destructive temperature and success of intensity determination, which suggests that this parameter have a potential as a pre-selection criterion. The new data double the Bulgarian archaeomagnetic dataset for the period in concern with 18 directional and 7 intensity results (6 full vectors). An inclination and intensity decrease is outlined at around 400 BC with assumed variation rates 0.10°/year and 0.32 μT/year, respectively. Considering the whole database, three intensity peaks are detected at circa 1050 BC (Fa ~ 78 μT), 550–500 BC (Fa ~ 96 μT) and 300 BC (Fa ~ 80 μT). The second peak is the highest over the entire eight-millennium long Bulgarian archaeomagnetic record. The first half of the Iron Age (1200–600 BC) is characterized with easterly declinations reaching ~18°E around 1150 BC, while in the last half it turns westerly with westernmost values around 400 BC. The current state of Bulgarian archaeomagnetic reference curves for the Early Iron Age is not sufficient for precise archaeomagnetic dating. Although the resulting dating interval for Ryahovets oven (the only structure with uncertain archaeological dating) coincides with the archaeological evidence, it spans more than 200 years. In such a case, correlation with other archaeomagnetically studied sites from the corresponding period can be especially helpful.