Loess-palaeosol sequences represent the most complete terrestrial archives of Quaternary climatic cyclicality. Particle size and geochemistry are widely used proxy data in palaeoclimatologic analysis of loess-palaeosol sequences. The palaeoclimatologic signals hidden in the texture and chemical composition of the Czech loess-palaeosol sequences, which are part of the European loess belt, are modulated by temporal changes in the interplay of oceanic and continental macroclimates and by the diverse bedrock geology of the Bohemian Massif.

Innovative tools of compositional data analysis, including log-ratio transformation and scalar-on-function regression, can substantially enhance the information value of large granulometric and geochemical datasets, when compared to classical statistics of raw data. Particle size distribution and bulk-rock geochemistry of 389 and 542 samples, respectively, from four Czech loess-palaeosol sequence sections representing the last glacial-interglacial cycle were analysed in this study. Centered log-ratio transformation was applied to the key elemental proxies and their spatial (between sections) and stratigraphic (within section) distribution. Centered log-ratio transformed densities of key elements were then plotted against particle size distribution to assess the particle size control on element concentrations.

Nearly all loess-palaeosol sequence samples exhibit a bimodal particle size distribution with medium/coarse silt and coarse clay fraction representing the main modes. The dominant silt fraction is completely allogenic. The clay component is partly allogenic, transported by high altitudes air stream, and partly authigenic (neoformed in situ by pedogenic processes). However there is only a minor variation in particle size between the sections, the centered log-ratio transformed density functions reveal that the particle size control on the distribution of major and trace elements is highly site-specific. The provenance signal is recorded especially in coarser-grained fractions transported for a short distance from the source area by near-surface wind. The proportion of the authigenic clay fraction, the alteration of feldspars, micas and low alkali contents indicated by high Rb/K ratio and high values of Rb/Sr and Sr/Ca ratios, which indicate the intensity of carbonate leaching, reflect the intensity of chemical weathering, which is highest in the podzols. Precipitation rates were likely the main microclimatic factor that controlled the compositional differences. The loess-palaeosol sequences in areas with higher present-day annual precipitation show higher contents of clay minerals, higher degree of loessification (cementation) and subsequent pedogenic changes.

黄土古土壤序列代表了第四纪气候周期性的最完整的陆地档案。粒度和地球化学是黄土-古土壤序列古气候分析中广泛使用的替代数据。隐藏在欧洲黄土带一部分的捷克黄土古土壤序列的质地和化学成分中的古气候信号受海洋和大陆宏观气候相互作用的时间变化以及波西米亚地块的不同基岩地质学的影响。

与原始数据的经典统计相比，组成数据分析的创新工具，包括对数比转换和函数标量回归，可以大大提高大型粒度和地球化学数据集的信息价值。在这项研究中，分别分析了来自四个捷克黄土-古土壤序列剖面的389个和542个样品的粒度分布和块岩地球化学，这些剖面代表了最后的冰河-冰期循环。将中心对数比转换应用于关键要素代理及其空间（剖面之间）和地层（剖面内）分布。然后将关键元素的中心对数比转化密度与粒径分布作图，以评估对元素浓度的粒径控制。

几乎所有的黄土-古土壤序列样品均表现出双峰粒度分布，其中中等/粗粉砂和粗粘土含量代表主要模式​​。占主导地位的淤泥部分是完全异源的。粘土成分部分是同种异体的，通过高海拔空气流运输，部分是自生的（通过成岩作用原位成岩）。但是，各部分之间的粒径只有很小的变化，居中的对数比转换的密度函数表明，对主要元素和痕量元素分布的粒径控制具有很高的位点特异性。出处信号尤其是在近地风将其从源区域转移到短距离的较粗粒级分中记录的。自生粘土成分的比例，长石的变化，较高的Rb / K比和高的Rb / Sr和Sr / Ca比值表示云母和低碱含量，这表明碳酸盐浸出强度反映了化学风化的强度，这在土壤中是最高的。降水率可能是控制组成差异的主要微气候因素。当今年降水量较高的地区的黄土古土壤序列显示粘土矿物质含量较高，黄土化程度（胶结）程度较高以及随后的成岩作用变化。降水率可能是控制组成差异的主要微气候因素。当今年降水量较高的地区的黄土古土壤序列显示粘土矿物质含量较高，黄土化程度（胶结）程度较高以及随后的成岩作用变化。降水率可能是控制组成差异的主要微气候因素。当今年降水量较高的地区的黄土古土壤序列显示粘土矿物质含量较高，黄土化程度（胶结）程度较高以及随后的成岩作用变化。