Quantitative estimates of Holocene precipitation from aeolian sand-palaeosol sequences across the Ordos Plateau, northern China, based on surface soil geochemistry

Abstract

Quantitative precipitation reconstructions are important for studies of past Earth surface processes and for predicting future trends, especially in the semi-arid region of northern China, where the dune fields landscape is dominated by shifts of the Asian summer monsoonal boundary. However, the relationships between desert deposits and climatic factors are controversial and it remains unclear how these relationships can be used for quantitative paleoenvironmental reconstruction. We investigated the relationship between the degree of chemical weathering of surface soil, quantified using the Chemical Index of Alteration (CIA) and modern climatic variables, along transects across the Ordos Plateau. The relationship was then used to obtain quantitative estimates of Holocene mean annual precipitation (MAP) from aeolian sand-palaeosol sequences in this region. The results reveal that MAP is the dominant climatic variable controlling the chemical weathering of surface soil, accounting for ∼ 69–91% of the total explained variance. This robust relationship enabled us to establish a regional CIA-based transfer function to reconstruct the MAP. This relationship was confirmed by a study of Holocene climate change and chemical weathering of aeolian sequences across the Ordos Plateau, which indicated that paleo-precipitation explained ∼ 76–88% of the total explained variance. The reconstruction demonstrates that the MAP was respectively ∼ 60–103 mm and ∼ 21–63 mm lower in the early and late Holocene than the modern MAP, while it was ∼ 34–77 mm (∼20–30%) higher in the middle Holocene than the modern MAP. Our results also demonstrate the latitudinal zonality of MAP evolution across the Ordos Plateau during the mid-Holocene. Overall, our findings provide new insights into the chemical weathering of surface soils and the response of Holocene aeolian deposits to climate change, and they confirm the utility of a geochemical proxy for quantitative environmental reconstruction in the semi-arid desert region of northern China.