[M-IS15] グローバル南極学

[MIS15-P10] Challenges in modeling daily-scale variability of stable isotopes in water vapor over the Southern Ocean

*木野 佳音¹、岡崎 淳史²、Bong Hayoung³、芳村 圭⁴ (1.東京大学大学院工学系研究科社会基盤学専攻、2.千葉大学国際高等研究基幹、3.NASA Goddard Institute for Space Studies、4.東京大学生産技術研究所)

キーワード：南極、気候モデル、大気の川、水同位体

Stable water isotope ratios (δ¹⁸O and δD) are effective tracers for the hydrological cycle due to their sensitivity to phase changes. Water isotope records from Antarctic ice cores are widely used as proxies for past climate variations. However, uncertainties remain regarding which processes are reflected in the isotopic signals preserved in ice cores. A recent study (Kino et al., 2021) has shown that an isotope-enabled climate model (isoGCM) fails to adequately reproduce the daily variations in precipitation isotope ratios over Antarctica during austral summer. It remains unclear whether this poor reproducibility is due to issues in simulating water vapor isotope ratios over the Southern Ocean or problems in representing transport processes over the Antarctic continent. As a first step toward addressing this issue, we evaluated whether isoGCMs can accurately reproduce the observed daily variations in water vapor isotope ratios over the Southern Ocean during austral summer. Focusing on the Indian Ocean sector of the Southern Ocean, this study compared isoGCMs (IsoGSM and MIROC5-iso; Bong et al., 2024) with ship-based observational data in January 2006 (Uemura et al., 2008). Although the models accurately reproduced the observed temperature, pressure, and humidity, they faced challenges in replicating the water vapor isotopic ratios. By categorizing atmospheric conditions into southerly cold and dry airflows from Antarctica, northerly flows accompanying atmospheric rivers, and other remaining cases, we found that the model–observation consistencies differed among the cases. In the southerly flow cases, the models failed to represent observed excessive decreases in δD and δ¹⁸O, while in the northerly cases, the models tend to overestimate the decreases in δD and δ¹⁸O. Nevertheless, simulated d-excess (deuterium excess, defined as δD – 8×δ¹⁸O) was comparable with the observation during the whole observation period. In order to interpret these overall discrepancies, we developed a new simple isotope model. The details of this model and its implications for understanding the problems in isoGCMs will be presented and discussed.