Middle-Late Holocene climate change in an East Asian sandy land: Perspectives from the phytolith record of different geomorphic units.

• Synthesizing the climate signal from different geomorphic units can better represent regional climate. • The local climate signals from different geomorphic units respond differently to regional climate due to micro-topography. • The periodicities in regional climate signals are mainly affected by solar activities. The accuracy of the regional palaeoclimate reconstruction is vital for ecological restoration and future climate predictions. Clarifying the regional climate signal on different geomorphological units within the same region plays a key role in regional climate researches. Here, we select the Horqin Sandy Land as the research area, and collect sedimentary from two typical geomorphological units (interdune lowlands: ML profile, DBS profile; dunes: CS profile, FX profile) as the research objects. The regional averaging method is employed to isolate distinct climate signals originating from interdune lowlands (IC-Score (Peat)) and dunes (IC-Score (Dune)), alongside an amalgamated climate signal (IC-Score (Peat&Dune)) synthesized from these two geomorphological units. This methodological approach facilitates the identification of variations in climate signals that collectively encapsulate the broader regional context. Meanwhile, local microclimate signals from dunes (IC-D-Value (Dune)) and interdune lowlands (IC-D-Value (Peat)) were separately analyzed to examine microclimate changes on different geomorphological units. The results indicate that different sedimentary carriers on various geomorphological units capture the local climate characteristics of their own units while recording regional climate changes. IC-D-Value (Peat) reflects 'warm-humid' climate, and IC-D-Value (Dune) represents 'cold-dry' climate. Climate signals derived from diverse geomorphological units exhibit the capacity to portray overarching regional climate patterns, yet they manifest variations concerning both semi-quantitative and quantitative climate parameters, notably temperature and precipitation. A regional average method is used to synthesize climate signals from different geomorphological units, which is more advantageous in documenting regional climate change. Regional climate signal reveals a change from warm-humid (9300 ∼ 4000 cal yr BP) to cold-dry (4000 ∼ 2000 cal yr BP) and then to slightly warm-humid (2000 ∼ 560 cal yr BP). Furthermore, regional climate signal shows the significant ∼ 2000-yr，∼1000-yr, ∼500-yr periodicities since 7300 cal yr BP, and solar activities are identified as one of the important driving factors. [ABSTRACT FROM AUTHOR]