No abstract

[1] This study aims to establish a stacked climate record of the Quaternary period from the Chinese loess sequence and to address the forcing mechanisms for the regional climate history of the Loess Plateau by correlating the stacked record with a composite d 18 O record in deep-sea sediments. A total of 18,352 samples were obtained from five loess sections, located at Baoji, Lingtai, Jingchuan, Puxian, and Pingliang in the southern and middle Loess Plateau. These yielded high-resolution grain size records. Between-section correlation of these shows that although small depositional hiatuses are present in places within a single section, most parts of the sections display near-continuous dust deposition throughout the Quaternary. The grain size records were tuned simultaneously to the theoretical variations in obliquity and precession of the Earth's orbit. The grain size records plotted on their orbital timescales were then averaged to form a stacked loess grain size time series, termed the ''Chiloparts'' record. This resolves most of the orbital timescale paleoclimate events buried in the loess-soil sequences of the southern and middle Loess Plateau and can be used as a regional archive of the Pleistocene climate history in the Loess Plateau. Comparison of the ''Chiloparts'' record with a composite marine d 18 O record shows that for the past 1.8 Ma, the loess-paleosol record can be correlated almost cycle by cycle with the marine record. Several discrepancies in the climatic events between the two records have also been identified, implying that regional forcing mechanisms may have played a part in the climatic evolution of the Chinese Loess Plateau.

No abstract

Real-world data are often acquired as a collection of matrices rather than as a single matrix. Such multiblock data are naturally linked and typically share some common features while at the same time exhibiting their own individual features, reflecting the underlying data generation mechanisms. To exploit the linked nature of data, we propose a new framework for common and individual feature extraction (CIFE) which identifies and separates the common and individual features from the multiblock data. Two efficient algorithms termed common orthogonal basis extraction (COBE) are proposed to extract common basis is shared by all data, independent on whether the number of common components is known beforehand. Feature extraction is then performed on the common and individual subspaces separately, by incorporating dimensionality reduction and blind source separation techniques. Comprehensive experimental results on both the synthetic and real-world data demonstrate significant advantages of the proposed CIFE method in comparison with the state-of-the-art.

The major objective of this study is to determine the spatial-temporal relationship between eolian sedimentation rate and particle size in the Loess Plateau, based on six highresolution 1oess records. The linkage of aridity changes in 1oess source areas to the winter monsoon system will also be addressed.