Variation of archaeomagnetic intensity since 1500 BC

Palaeomagnetic measurements of heated archaeological structures are able to resolve the Earth`s magnetic field`s secular variation on timescales in the order of several decades. This allows the prolongation of spherical harmonic models which have been obtained from his-torical direct measurements of the magnetic field for the past 400 years to a time interval of several millennia. Crucial for such modeling and its temporal resolution is a sufficiently dense data base, especially for archaeointensity. While Western, Central and Southern Europe are well covered with a directional data set for at least the past 2000 years, the amount of inten-sity data is still poor and values show a lot of scatter. Sufficiently dense and precise data cover only the past 700 years. The main goal of the project is to carry out archaeointensity measurements on archaeological sites with known archaeodirection. It is intended to enhance for the past 3500 years the archaeomagnetic full vector data set of Central Europe from one or two values per century to about five full archaeomagnetic vectors. This aim will be achieved by first testing and then applying a new method for palaeointensity determination which is able to avoid disturbing chemical alterations and allows for correction of domain effects. Compared to the classical Thellier method it requires only about one third of the heating steps. Therefore, palaeointensity determination is much quicker. The new method also can use material which fails to provide results with the Thellier method. Obtaining such a dense full vector archaeomagnetic data set will provide the data base for improved modeling of the magnetic field during the past 3500 years and increase precision and scope of archaeo-magnetic dating. In this time interval five `archaeomagnetic jerks` occurred defined by sudden changes in the movement of the archaeodirection. Only one of them, around 800 AD seems to be correlated with significantly enhanced archaeointensity. The archaeomagnetic jerk at 900 BC is characterized by a large and rapid loop the Eastern declinations up to 70 and positions of the virtual geomagnetic poles at relatively low latitudes around or less than 60N. At the same time the archaeointensity decreases to about 2/3 of the present value followed by a rapid increase to the double value. These features are documented in several sites in Austria, Ger-many and France, but require further confirmation. The strong increase of archaeointensity lasting from 800 to 500 BC is correlated with increasing rainfall and a change to colder climate all over Europe. The same is true for the archaeointensity maximum found around 800 AD. Both time intervals are also correlated with lower rates of radiocarbon production which lead to long plateaus in the radiocarbon calibration curve. Accordingly, links between the Earth`s magnetic field and processes in the high atmosphere governing radionuclide or cloud production become more obvious. However, an increasing archaeomagnetic data base for better constraining of global magnetic field models is still needed.

The projects results contribute to two different scientific fields. On one hand, the geophysical aspect is that a considerably enlarged data set of geomagnetic vectors of the time interval 1500 BC to 2003 AD provides a better knowledge of geomagnetic field variations for the past 3500 years. This data set overlaps with direct historical observations of the geomagnetic field. This allows for access to the reliability of so-called archaeomagnetic measurements, which retrieve information on the geomagnetic field vector by investigation of any kind of ovens, baked clay, bricks and pottery found in archaeological excavations. On the other hand, a soon as the temporal variation of the geomagnetic field is known for a certain region, it can be used to determine the age of other, undated archaeological features containing baked material. This is a well established method when the direction of the geomagnetic field can be used. The challenge of the project was to increase the data set on geomagnetic field strength, which is much more difficult to measure. For this purpose a new method was tested on a large number of archaeological features. Although this new method is not much quicker or more successful than established methods, it has the advantage to provide smaller errors and it gives a better access on reliability of the data. The newly obtained data further support that very strong variations of the archaeomagnetic field strength (strong increase followed by a decrease) occurred in two time intervals around 800 BC and 800 AD, each lasting several centuries. Especially for these time intervals also new directional data have been obtained in the project and help to define the geomagnetic field variations much better. They show that also strong movements in geomagnetic direction occurred in these two periods. Both time intervals are coincident with long plateaus of the radiocarbon calibration curve and hence very imprecise 14C-ages are systematically obtained in these two important archaeological periods, which are the transition from Bronze to Iron Age and the early Medieval. The strong geomagnetic variations are now sufficiently well characterized, that they can be used for archaeomagnetic dating in Central Europe and help to improve archaeological chronologies.The data are as well important for the geophysical aspect of geomagnetic field modeling. Crucial for such modeling and its temporal resolution is a sufficiently dense data base, which due to the projects results is now available for Central Europe for the first time. This will lead to a better understanding of strong geomagnetic variations on time scales of decades to millennia.