Secular variation between 5 and 10c CE in Japan: remeasurements of 2000 samples collected between 1960-70's from Sueki earthenware kilns in Osaka.

Abstract:

In Japan, archeomagnetic measurements are vigorously developed for years, though it is not well known to paleomagnetism community in english. One of the works is massive archeomagnetic study of Suemura kiln group carried out in Osaka University in 1960's to early 70's. More than 500 kilns were excavated in Sakai city and vicinities, Osaka Prefecture, Japan. The kiln group is called as Suemura Kilns, and are for Sueki earthenware of 5c to 10c CE. About 300 kilns were sampled and most of the samples were measured at the time, and the results are reported in e.g. Hirooka (1971) and Shibuya (1980). However, the results have significant scatter in direction, which may be due to the limitation of old astatic magnetometer measurements and handwriting graphic determination of magnetic direction, and/or the lack of demagnetization. We recently inherited many of those samples and remeasured them with spinner magnetometer applying alternation field demagnetization (afd). The magnetizations are generally very stable, as usual as other archeomagnetic samples, and afd does not change the magnetic direction mostly. However, significant number of sites show large scatter in magnetic directions, which might be due to the wrong identification of kiln floor or disturbance at the time of collapsing or excavation. Taking kilns of α₉₅<4^o, we recovered 131 paleomagnetic directions. Although third of them are dated by pottery shape chronology, the range of each pottery style is not precisely known and the relation of the baked floor and the potteries excavated around kilns are not always clear. The carbon dating of those kilns are very scares. Thus we first try to draw secular variation curve in declination-inclination plot. With the rough ages of those kilns, it is pretty easy to draw the SVC. It is also numerically determined taking the distance of each direction from nearest point in SVC and the velocity change of the SVC as penalty function, within a couple of degrees in the error. The the age of each point is assigned to satisfy the archeological ages.

This precise SVC in the far east will improve understanding the geomagnetic variations, as well as application to the archeological dating.