Seasonal and Hydroclimatic Dynamics Revealed by δDwax and δ13Cwax in a Tropical Lake Core for the past 18 000 years

Leaf wax hydrogen isotopic composition (δD_wax) is increasingly used to reconstruct precipitation variability over geological time scales. A significant advantage of δD_wax analysis compared to isotopic proxies from speleothems or ice cores is that the isotopic composition of a range of n-alkanes sourced from algae, aquatic plants and terrestrial plants is measured simultaneously. This additional data allows for investigations of multiple aspects of the hydrological cycle such as dry season evapotranspiration and seasonal distribution of rainfall. However, uncertainties persist in interpreting the δD of different n-alkanes, and as a result, δD_wax studies often focus on a single terrestrial n-alkane (C₂₇, C₂₉, or C₃₁). The choice of n-alkane can significantly influence the reconstructed precipitation record, potentially leading to divergent interpretations. Moreover, only a subset of the available data is typically utilized, suggesting that leaf wax analyses may hold untapped potential for deeper insights. 

In this ongoing study, we explore the n-alkane distribution and hydrogen and carbon (δ¹³C_wax) isotopic composition from an 18 000-year-long lake core in Thailand. High-resolution radiocarbon dating (n=65) was used to construct a well-constrained age model for the deglaciation. Our results indicate that δD_wax records intense wetting during the Bølling and Older Dryas periods (~14.7-13.9 ka BP). Interestingly, δ¹³C_wax reveals a simultaneous shift toward dry-adapted C4 plants, suggesting seasonally dry conditions that impact vegetation, while δD_wax likely reflects a wet-season signal. Allerød (13.9-12.8 ka BP) is marked by drying coinciding with a shift back to wet-adapted C3 vegetation. δD_wax thereby appears to be seasonally biased in this tropical lake, at least for the deglacial period. The hydroclimate response during the Younger Dryas cold period (12.8-11.7 ka BP) is ambiguous based on different n-alkanes in our δD_wax record – and even sign of change varies for the C₂₅ to C₃₁n-alkanes. Our results thereby highlight the need to carefully consider which n-alkane to use in hydroclimate δD_wax reconstructions, and we discuss what information these differences may provide. Specifically, the divergence between aquatic and terrestrial δD_wax may serve as a useful indicator of drought-induced evapotranspiration in this lacustrine setting.