| 引用本文: | 陈茂柏,唐玲,周卫健,孔祥辉,杜雅娟.2026.黄土磁化率衰减的发现和校正[J].地球环境学报,17(4):997-1008 |
| CHEN Maobai,TANG Ling,ZHOU Weijian,KONG Xianghui,DU Yajuan.2026.Discovery and correction of loess susceptibility decline relying on cosmogenic 10Be record in loess[J].Journal of Earth Environment,17(4):997-1008 |
|
| 摘要: |
| 数十年来，利用直接测量的黄土磁化率数据，国内外学者发表了大量来自我国黄土的古气候研究论文，从而使我国黄土-古土壤沉积序列跻身于全球变化研究三大支柱之一。然而，基于西峰黄土剖面0—870 ka宇宙成因核素10Be浓度曲线与磁化率曲线的高度相似性，利用创建的黄土10Be“残差示踪法”（RTA）计算得到的黄土10Be残差曲线及其重建的大气10Be产率曲线都显现出不正常的倾斜或拱起趋势，从而意外发现黄土磁化率可能存在衰减现象，以往的研究均未考虑这一重要问题。基于此，文章不仅通过计算确证了黄土磁化率衰退的事实，而且创建了磁化率的“反演μ校正”方法以恢复衰退的磁化率值；同时还采用“金标准间接检验”法来验证被校正磁化率的正确程度。μ校正结果表明西峰0—870 ka剖面黄土磁化率的平均衰退率高达87%。“金标准间接检验”表明检测曲线与被检验曲线几乎完全相似（r=0.99），两条比较曲线间点对点差异的相对标准偏差（RSD）为4.43%。黄土磁化率衰退的“发现”和“校正”是文章的两个创新点，黄土磁化率衰退的“机制”尚待进一步研究。 |
| 关键词: 黄土磁化率 衰减 校正 |
| DOI：10.7515/JEE2024126 |
| CSTR：32259.14.JEE2024126 |
| 分类号: |
| 文献标识码:A |
| 基金项目:国家自然科学基金重点项目（41930321）；中国科学院战略性先导科技专项（B类）（XDB1660100）；崂山实验室科技创新项目（LSKJ202203300）；陕西省创新能力支撑计划（2025SYS-SZSYS-40） |
| 英文基金项目: |
|
| Discovery and correction of loess susceptibility decline relying on cosmogenic 10Be record in loess |
|
CHEN Maobai1，2，TANG Ling3，4，ZHOU Weijian1，3，KONG Xianghui1，3，DU Yajuan1，3
|
|
1.State Key Laboratory of Loess Science, Institute of Earth Environment, Chinese Academy of Sciences, Xi’an 710061 , China ；2.Shanghai Institute of Applied Physics, Chinese Academy of Science, Shanghai 201800 , China ；3.Shaanxi Provincial Key Laboratory of Accelerator Mass Spectrometry Technology and Applications, Xi’an AMS Center, Xi’an 710061 , China ；4.Xi’an Institute for Innovative Earth Environment Research, Xi’an 710061 , China
|
| Abstract: |
Background, aim, and scope Attributed to the combination of the measured cosmogenic 10Be record in loess and the developed “Residual Tracing Approach” (RTA) mathematical trace method, decline of loess susceptibility in the Xifeng 0—870 ka profile is unexpectedly revealed by the linear tilting or nonlinear arching tendency of both the RTA-calculated residual curve and the RTA-reconstructed 0—870 ka atmospheric 10Be relative production rate (Pr) curve. Since such a decline fact has long been ignored by loess community in the numerous published papers based on measured loess susceptibility values without being aware of their decline for various researches during the past several decades, this work is aimed at not only certificating the decline fact, but also developing an inversion μ correction method to recover the declined susceptibility. Also, an indirect verification method is proposed to assess accuracy of the μ-corrected loess susceptibility. Materials and methods The method of this work is based on calculation. Both applied 10Be concentration Be(M)m (AMS-measured loess 10Be concentration) and susceptibility SUS(M)m (measured loess magnetic susceptibility) data are adopted from the Xifeng 0—870 ka loess profile with high resolution. These data are measured on 3 MV AMS device and a Bartington MS2 susceptibility meter respectively at Xi’an AMS Center. The errors of the AMS-measured Be(M)m are ≤3% and those of the measured SUS(M)m are ≤1%. Results The μ-correction has shown that the average decline amount of the Xifeng 0—870 ka loess susceptibility is as high as 87%. Indirect verification has indicated almost complete similarity (r=0.99) and acceptable relative standard deviation of point-to-point difference (RSD=4.43%) between the verifying curve related to the Golden Standard of Verification and the examined curve related to the recovered susceptibility. Discussion Although good similarities between decay-corrected Be(M)m curves and measured SUS(M)m curves for different Chinese loess profiles have been reported for more than thirty years, the loess community has always neglected such a simple fact that the loess susceptibility SUS(M)m should be synchronously declining with the radioactivity decayed 10Be concentration Be(M)m so as to be able just to keep their good correlation to each other all the time. Furthermore, owing to modulation by extremely inhomogeneous climate condition over Chinese Loess Plateau, this loess susceptibility decline was twisted with fluctuating climate effect and could not be detected by the paleomagnetic method. On the contrary, it is the linear regression between the decay-corrected Be(M)m and the measured SUS(M)m in the RTA method that has removed most (about 70%) climate influences on the decay-corrected Be(M)m by the measured climate proxy SUS(M). As a result, the susceptibility declines in the Xifeng 0—870 ka loess profile is revealed. Conclusions This work reveals the simple fact of loess susceptibility decline, and the decline situation of loess susceptibility would be varied for different profiles, even varied for the same loess site during different age intervals. The developed and verified μ correction method for the Xifeng 0—870 ka profile is one of the effective methods in practice for recovering the declined loess susceptibility, which is also appropriate to other loess profiles to obtain more accurate fitted linear/curve of the unlevel residual curve for the accurate residual correction amount ε. However, the Indirect Verification of Golden Standard of their corrected susceptibility is only suitable to such profile that the necessary and sufficient conditions (R  = R  ≈ 0.9 and the R >R2) must be available. Recommendations and perspectives Both discovery and correction of loess susceptibility decline are significant advancements in loess science. For example, the measured loess susceptibility cannot be taken as a climate proxy until it is corrected by using the correction method. Accordingly, many published papers based just on measured susceptibility might require re-evaluations. Further investigation into the decline mechanism of loess susceptibility is recommended. |
| Key words: loess susceptibility decline correction |
