Late Pleistocene-Holocene aeolian loess-paleosol sections in the Yellow River source area on the northeast Tibetan Plateau: chronostratigraphy, sediment provenance, and implications for paleoclimate reconstruction

Two late Pleistocene-Holocene aeolian loess-paleosol sections were identified in the Yellow River source area on the northeast (NE) Tibetan Plateau. Based on the comprehensive study on multiple proxies, AMS 14C and optically stimulated luminescence (OSL) dates, these two sections have a basal age of 14000–11500 yr BP, suggesting that they accumulated since the last deglaciation. Four stages of paleoclimate evolution can be summarized as follows. (1) A cold-dry period was dominated by the Westerlies in the late Pleistocene (ca. 14000–11500 yr BP). (2) A gradual transition to warm-wet with the weakened Westerlies and strengthened East Asian summer monsoon (EASM) occurred in the early Holocene (ca. 11500–9000 yr BP). (3) A warm-wet episode was under strong EASM in the middle Holocene (ca. 9000–3100 yr BP), however, a cold event was identified at 5000–3500 yr BP, which was caused by the abrupt weakening of the EASM and the strengthening of the Westerlies. (4) A phase of gradually shifting to dry-cold was controlled by the weakened EASM and strengthened Westerlies (after ca. 3100 yr BP). The assemblage of grain-size end-member modelling analysis (EM1-EM5) and heavy minerals indicated that aeolian deposit units have multiple sediment provenances and transport processes. EM1 is clay deposits composed of secondary clay minerals produced during weathering and pedogenesis, whereas EM2 is fine silts derived from distant dust transported through long-existing high-level Westerlies. EM3 and EM4 are coarse silts that originated from the moraines and periglacial deposits in the middle and distant source areas transported by the Westerlies and Tibetan Plateau Monsoon. Lastly, EM5 is sand from fluvial deposits and slope clastics transported by strong valley wind within the local area. This study is helpful to understand the mechanisms of paleoclimate evolution on the NE Tibetan Plateau. It also has important implications for characterizing and subdividing aeolian loess in China.