| As one of the important pillar of the global paleoclimatic change research, a high precision, high resolution timescale of loess in China is more and more important for quantitative research of paleoenvironmental chang.Up to now, Many researchers put forward to the establishment of different loess timescales, such as, linear interpolation model, magnetic susceptibility model, grain size model and orbtial tuning. However, there is no unified and accurate method to establish a timescale of loess in China. In these models mentioned above, which is more reliable and accurate? And Which grain size can be used as an excellent proxy index in grain size model? Answers of these questions will provide the reference to establish a high precision, high resolution timescale.The analyzed sequence here is the section of continuous eolian sediments collected from Changwu of ~1.2 Ma. Using magnetic susceptibility, grain size (median,<2μm size fraction,2-10μm size fraction,10-20μm size fraction, 20-32μm size fraction,>32μm size fraction,32-63μm size fraction and> 63μm size fraction) as the basic data,we establish differtent timescales by differtent time control points.The boundary ages of paleosoil which are based on marine istope stage (MIS) boundaries by comparing marine oxygen isotope curve and an integrate method of Blackman-Tukey power spectrμm analysis are both used to verifies accuracy of timescales.The major results are shown as follows:1) Verifying accuracy by two methods, timescale established by magnetic susceptibility model holds highest degree of accuracy.2) In different size fractions, accuracy of timescale establised by>32μm size fraction is the highest. Median size fraction’s is higher. |