New insights into loess formation on the southern margin of the Chinese Loess Plateau

2021 
Abstract The Loess Plateau has been identified as one of the most important areas for dust particles deposition and considered as one of the key areas for studying the global climate change and regional responses. In this study, a continuous loess profile, located on the southern margin of the Loess Plateau, was selected to analyze microstructure of the loess and palaeosol. Combined with the end-member model and principal factor analysis, the dust transport and loess formation processes were reconstructed. Magnetic susceptibility and grain-size analyses suggest considerable differences between the loess and the palaeosol. The parametric decomposition model was used to decompose the grain-size components into 3 end-members, revealing the transport force of westerly circulation and winter monsoon and pedogenesis after deposition that affected the loess formation. The microstructure analysis reflects the loess is different from the palaeosol in terms of skeleton particles, contact modes, type and morphology of pores. More meso- and minipores and smaller fractal dimension of pores is observed in loess than palaeosol, preserving the climatic information during the formation of loess layers. The principal factor analysis of grain-size components and parameters shows that factor Ⅰ indicates the two important transport force of westerly circulation and winter monsoon. While factor Ⅱ reflects a kind of compound force, the combined effect of high airflow transport and pedogenesis. Based on these, the conceptual model of loess formation in the Weihe Basin is established under different climates between glacial and interglacial. The findings are helpful to understand the regional climatic environmental evolution, formation of loess landform and evolution of the atmospheric circulation system in East Asia.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    97
    References
    0
    Citations
    NaN
    KQI
    []