Paleomagnetic and rock magnetic studies of the Sakurajima 1914 and 1946 andesitic lavas from Japan: A comparison of the LTD-DHT Shaw and Thellier paleointensity methods
- 作者：Yuhji Yamamoto ; Hiroyuki Hoshi
- 关键词：Paleointensity ; Thellier method ; LTD-DHT Shaw method ; Sakurajima
- 刊名：Physics of the Earth and Planetary Interiors
- 出版年：2008
- 期刊代码：73_00319201
- 类别：ear
- 出版时间：March 2008
- 卷：167
- 期：1-2
- 页码：118-143
- 文件大小：3826 K
摘要
Paleomagnetic and rock magnetic measurements have been made on samples from the 1914 and 1946 Sakurajima andesitic lava flows in order to investigate the suitability of andesite to record both paleodirection and paleointensity. The lava flows were split in to three sites, each of which consists of two or three sub-sites. A range of rock magnetic experiments have been made on multiple samples per flow and the samples can be split into two broad rock magnetic categories (I and II). Paleodirection was determined by stepwise thermal demagnetization and the mean direction determined from both flows is very close to that expected from IGRF-10. Paleointensity was determined using the LTD-DHT Shaw technique and the Thellier method on sister samples of the same cores. Sixty four out of 72 LTD-DHT Shaw paleointensity experiments were successful and there are no differences between the sub-sites. Seventy out of 71 Thellier paleointensity experiments were successful; however, for sub-sites in two out of the three sites there are large differences between the mean paleointensities. When all paleointensities are normalized to the expected value from IGRF-10, the mean is 0.98±0.11 (±σ) for the LTD-DHT Shaw method and 1.12±0.13 (±σ) for the Thellier method. These means are statistically distinguishable and this is related to the different rock magnetic properties of the samples. If only category I samples are considered, the LTD-DHT Shaw mean is 1.05±0.07 compared with the Thellier mean of 1.08±0.08. In contrast there is a much larger spread in the means determined from category II samples: 0.94±0.11 from the LTD-DHT Shaw experiments and 1.15±0.15 from the Thellier experiments. This spread could be caused by magnetic interactions between magnetic domains. Assuming high temperature oxidation of titanomagnetite grains during natural cooling, we suggest a scenario that can qualitatively explain the difference between the LTD-DHT Shaw and Thellier paleointensity results.