[M-IS06] Evolution and variability of the Tropical Monsoon and Indo-Pacific climate during the Cenozoic Era

[MIS06-09] Reconstructing past dynamics of the East Asian Monsoon using the temperature variability derived from individual foraminifera analysis

*Pei-Ting Lee¹、Sze Ling Ho¹、Yoshimi Kubota²、Keiji Horikawa³、Chuan-Chou Shen⁴、Min-Te Chen⁵ (1.Institute of Oceanography, National Taiwan University, Taipei, Taiwan、2.Geology and Paleontology, National Museum of Nature and Science, Tsukuba, Japan、3.Faculty of Science, Academic Assembly, University of Toyama, Toyama, Japan、4.Department of Geosciences, National Taiwan University, Taipei, Taiwan、5.Institute of Earth Sciences, National Taiwan Ocean University, Keelung, Taiwan)

キーワード：Individual foraminifera analysis, Mg/Ca, Oxygen isotope ratio δ18O, Mixed layer thickness

The East Asian Monsoon (EAM) system exerts strong control over the surface circulation and mixing in the upper ocean of the South China Sea (SCS). As such, upper water column mixing is often used to investigate the past dynamics of the EAM. One common approach is through the vertical temperature difference (ΔT) of foraminifera species inhabiting the mixed layer and thermocline, assuming constant habitat depths over time. However, this assumption contradicts modern observations indicating habitat depths vary in response to seasonal hydrographic changes. For instance, Trilobatus sacculifer, which dwells in the mixed layer of the SCS, tends to occupy a wider habitat depth range when the water column is stratified, in contrast to its narrower vertical distribution during monsoon months. In this study, we leverage the ecological response of T. sacculifer to infer past dynamics of the EAM. To validate this approach, we analyze the Mg/Ca values of individual T. sacculifer tests (hereafter referred to as IFA-Mg/Ca) from sediment trap samples spanning 2017-2019. Our results indicate that T. sacculifer IFA-Mg/Ca temperature distributions are predominantly unimodal during winter monsoon months but exhibit bimodal patterns during summer months, suggesting seasonal shifts in habitat depth. The bimodal summer pattern likely reflects the proliferation of T. sacculifer at the deep chlorophyll maximum when the mixed layer shoals in summer. Moreover, T. sacculifer IFA-Mg/Ca temperature variability (i.e., the spread in sample distribution) exhibits a strong negative correlation with the mixed-layer thickness. This result can be replicated using IFA-Mg/Ca data generated in a different laboratory and IFA-δ¹⁸O data, demonstrating its robustness as an indicator of upper water column mixing. Next, we use this approach to infer upper ocean mixing in the northern SCS (site MD97-2146) over the last glacial cycle as a proxy for EAM glacial-interglacial variability. Our results reveal that during the Holocene, T. sacculifer IFA-Mg/Ca temperature distributions displayed bimodal or trapezoidal patterns, whereas during the Last Glacial Maximum (LGM), they predominantly exhibited unimodal distributions, resembling modern seasonal patterns. The data indicate reduced T. sacculifer IFA-Mg/Ca temperature and IFA-δ¹⁸O variability during the LGM, alongside increased variability during the Holocene, suggesting more mixing and a thicker mixed layer during the LGM. These inferences are consistent with findings from land-based reconstructions. Conversely, the Mg/Ca temperature evolutions of thermocline-dwelling species Neogloboquadrina dutertrei and Pulleniatina obliquiloculata derived from conventional multi-specimen analysis, differ from one another. This discrepancy may stem from habitat depth shifts in one or both species. If true, this would violate the constant habitat depth assumption required for the application of the conventional ΔT approach. In summary, our findings suggest that IFA-Mg/Ca and δ¹⁸O variability, which reflect the ecological response of foraminifera to hydrographic changes, may be sensitive indicators for tracking past changes in EAM variability. Future research should investigate whether this approach is applicable to other regions.