An extended and revised Lake Suigetsu varve chronology from ∼50 to ∼10 ka BP based on detailed sediment micro-facies analyses
An extended and revised Lake Suigetsu varve chronology from ∼50 to ∼10 ka BP based on detailed sediment micro-facies analyses
复制标题
DOI：
10.1016/j.quascirev.2018.09.021
复制
发表时间：
2018-11
影响因子：
4
通讯作者：
G. Schlolaut;R. Staff;A. Brauer;H. Lamb;M. Marshall;C. Bronk Ramsey;T. Nakagawa
中科院分区：
文献类型：
--
作者：
G. Schlolaut;R. Staff;A. Brauer;H. Lamb;M. Marshall;C. Bronk Ramsey;T. Nakagawa
水月湖（日本）是东亚放射性碳（14 C）校准和古环境重建的关键地点。在这里，我们提出了一个描述的沉积物（微）相，结合一种新的方法来纹状插值允许建设一个修订的纹状年表，延长以前的2012年纹状年表的100万年，回到150万年BP。在纹层计数和插值方面的挑战，以前只详细讨论了末次冰期-间冰期过渡，这里描述了回到150 ka BP。此外，还讨论了μXRF扫描法和薄片显微镜法的相对优点。相分析揭示了四个相带，它们的转变受当地和气候控制。沉积物的分层质量是高度可变的，纹状插值显示，在分析的时间间隔内，平均而言，只有50%的年度周期是由季节性层。在剩下的几年里，季节层是难以区分的，即要么没有形成，要么没有保存下来。对于纹层插值，使用了高级版本的纹层插值程序，这使得能够构建最长的，纯粹的纹层年代学出版，尽管层压质量差的长时间间隔。计算出的插值不确定度为+8.9%和-4.6%，考虑到高插值度和记录长度，这完全在预期范围内。
Lake Suigetsu (Japan) is a key site for radiocarbon (14C) calibration and palaeo-environmental reconstruction in East Asia. Here we present a description of the sediment (micro)facies, which in combination with a new approach to varve interpolation allows construction of a revised varve based chronology that extends the previous 2012 varve based chronology by ∼10 ka, back to ∼50 ka BP. Challenges in varve counting and interpolation, which were previously discussed in detail only for the Last Glacial-Interglacial Transition, are described here back to ∼50 ka BP. Furthermore, the relative merits of varve counting by μXRF scanning and by thin-section microscopy are discussed. Facies analysis reveals four facies zones, their transitions driven by both local and climatic controls. The lamination quality of the sediment is highly variable and varve interpolation reveals that in the analysed time interval, on average, only 50% of the annual cycles are represented by seasonal layers. In the remaining years seasonal layers are indistinguishable, i.e. either did not form or were not preserved. For varve interpolation an advanced version of the Varve Interpolation Program was used, which enabled the construction of the longest, purely varve dated chronology published, despite long intervals of poor lamination quality. The calculated interpolation uncertainty is +8.9% and −4.6%, which is well within expectations considering the high degree of interpolation and the length of the record.