Climate change affects ecosystems and landscapes through changes in mean conditions (e.g., slow solar insolation change) (Joshi et al., 2011;Steinbauer et al., 2018) and through rapid climatic events (Alley et al., 2003). A recent study has emphasized that it is critical to shift focus from target states to target rates of ecological change, distinguishing fast, slow and abrupt responses to climatic forcing (Williams et al., 2021), where abrupt responses are nonlinear, and related to thresholds and tipping points (Lenton et al., 2008). Past climate change, especially during the Late Quaternary with frequent abrupt climatic oscillations, provides an opportunity to identify hotspots of vulnerability and resilience and determine management interventions. In order to differentiate the fast-slow-abrupt response rates, however, one of the most critical prerequisites is the availability of well-dated climatic records for analysis of leads and lags, especially in places sensitive to climatic change like the Tibetan Plateau (TP).The TP (Figure 1a) is a current hotspot of Earth system research, and the conservation status of the TP loess steppe ecoregion is expected to be vulnerable to changes in future warming scenarios (Beaumont et al., 2011). As one of the most sensitive areas to response to global climate change, warming in the TP occurred not only earlier based on modern observations (Liu & Chen, 2000), but also with large magnitude of temperature shifts based on climate modelling (Brown et al., 2020). Without understanding the processes and mechanisms of landscape and ecosystem evolution within the TP, it is challenging to adapt to and mitigate the consequences from