Astronomical forcing of lake-level and environmental changes in saline-alkaline lakes during the Late Paleozoic Ice Age

The Late Paleozoic Ice Age (LPIA) records Earth's latest icehouse-greenhouse transition, which offers a deep-time perspective for future climatic and environmental predictions. Astronomical cycles are believed to play a key role in the evolution of climatic and environmental changes during the LPIA. However, the interactions among the astronomical forcing, climatic conditions, sea-level and lake-level changes remain ambiguous. Here, we conducted a cyclostratigraphic analysis of gamma ray (GR) and paleo-water depth (WD) data of the upper Carboniferous–lower Permian Fengcheng and Lucaogou formations in the Junggar Basin, NW China. A floating astronomical time scale of ∼3.4 Myr for the Fengcheng Formation and ∼5 Myr for the Lucaogou Formation are established by astronomical tuning of GR data to the robust 405 kyr eccentricity cycle. The lake-level changes of the two formations are recovered based on the sedimentary noise model, which both show antiphase relationship with relative sea-level fluctuations, however different phase relationship with obliquity amplitude variations. The ∼1.2 Myr (s₄–s₃) obliquity amplitude modulation cycle was the main driving force of lake levels in the Northern Hemisphere during the LPIA. In addition, the ∼170 kyr cycles regulates the lake-level fluctuations and affect the development of saline-alkaline minerals in the closed lake basins. The cyclical growth and retreat of extensive ice sheets was an important factor that influence the hydrological response to the same obliquity cycle, which reflect the nonlinear response of Earth's climate system to changes in astronomical orbital parameters. We conclude that the combination of climatic background and astronomical forcing cannot be generalized, as they have a more complex superposition mechanism. Our study demonstrates the important role of the termination of icehouse climate in lake evolution and supports the existence of the 'nonlinear response process' across different time scales.