10:00 AM - 10:15 AM
[MIS20-05] Impacts of Intermittent Precipitation Events on Reconstructed Last Glacial Maximum Surface Temperature from Water Isotope Signals in Dome Fuji Ice Cores
Keywords:ice core, water isotopes, Antarctica, climate model, blocking event, Last Glacial Maximum (LGM)
Stable water isotope ratios (like H218O/H216O, expressed in conventional δ value, i.e., δ18O) obtained from Antarctic ice cores are widely used to reconstruct temperature variations in the past, like during the Last Glacial Maximum (LGM) ~21 000 years ago. The ice core studies, including Rayleigh-type models used to reconstruct past temperature changes from water isotope signals, relied on an implicit assumption for decades: a constant relationship between mean condensation temperatures and surface air temperatures (SATs) in inland Antarctica (e.g., Masson-Delmotte et al., 2010). However, recent observations of in-situ and satellites suggested that episodic precipitation events should bias those isotope signals, suggesting potential uncertainty in SAT reconstructions with this implicit assumption (e.g., Turner et al., 2019). With the isotope-enabled atmosphere general circulation model (AGCM) MIROC5-iso (Okazaki & Yoshimura, 2019), Kino et al. (2021) showed that a warm bias, influenced by daily scale episodic precipitation events, should exist in the austral winter for the present climate. They also showed that the Southern Annular Mode (SAM; an indicator for the westerly wind) characterizes daily scale episodic precipitation events, and even modifies the daily δ18Op/SAT relationship during the austral winter at Dome Fuji. However, the episodic precipitation events and their influence on the δ18Op/SAT relationship may change in seasonality and frequency for different climate states like the LGM.
In this study, simulations under LGM climate conditions were performed with MIROC5-iso using two independent sea surface boundary field conditions (sea surface temperature (SST) and sea ice fraction) (Paul et al., 2021; Sherriff-Tadano et al., submitted). We confirmed that the important processes controlling δ18Op at Dome Fuji were common among present and LGM climates, including the relationship with SAM. It suggests that the implicit assumption should be reconsidered for LGM, too. The spatial distribution of simulated δ18Op in precipitation over Antarctica differed in our 2 LGM simulations, especially around Dome Fuji (~30 %). The differences in SST gradient in the southern mid-latitudes primarily altered the SAT and δ18Op at Dome Fuji through the mean state of atmospheric circulation changes. The sea ice extension affected the environment of the water vapor transport pathway to Dome Fuji and enhanced the difference in δ18Op at the station, induced by the SST gradient changes. The non-linear effect of SAM on LGM-present δ18Op anomalies at Dome Fuji was secondary. On the other hand, the SST gradient altered the characteristic of SAM and the daily variations of LGM SAT and δ18Op at Dome Fuji. We estimated an uncertainty induced by the implicit assumption of 58–129 % for the LGM to present changes in SAT. Further LGM sensitivity simulations with isotope-enabled AGCMs would be necessary to constrain this uncertainty.
[1] Kino et al., J. Geophys. Res., 126(23). https://doi.org/10.1029/2021jd035397, 2021.
[2] Masson-Delmotte et al., Quat. Sci. Rev., 29(1), 113–128, https://doi.org/10.1016/j.quascirev.2009.09.030, 2010.
[3] Okazaki & Yoshimura, J. Geophys. Res., 124(16), 8972–8993. https://doi.org/10.1029/2018jd029463, 2019.
[4] Paul et al., Clim. Past, 17(2), 805–824. https://doi.org/10.5194/cp-17-805-2021, 2021.
[5] Sherriff-Tadano et al., submitted.
[6] Turner et al., Geophys. Res. Lett., 46(6), 3502–3511. https://doi.org/10.1029/2018gl081517, 2019.
In this study, simulations under LGM climate conditions were performed with MIROC5-iso using two independent sea surface boundary field conditions (sea surface temperature (SST) and sea ice fraction) (Paul et al., 2021; Sherriff-Tadano et al., submitted). We confirmed that the important processes controlling δ18Op at Dome Fuji were common among present and LGM climates, including the relationship with SAM. It suggests that the implicit assumption should be reconsidered for LGM, too. The spatial distribution of simulated δ18Op in precipitation over Antarctica differed in our 2 LGM simulations, especially around Dome Fuji (~30 %). The differences in SST gradient in the southern mid-latitudes primarily altered the SAT and δ18Op at Dome Fuji through the mean state of atmospheric circulation changes. The sea ice extension affected the environment of the water vapor transport pathway to Dome Fuji and enhanced the difference in δ18Op at the station, induced by the SST gradient changes. The non-linear effect of SAM on LGM-present δ18Op anomalies at Dome Fuji was secondary. On the other hand, the SST gradient altered the characteristic of SAM and the daily variations of LGM SAT and δ18Op at Dome Fuji. We estimated an uncertainty induced by the implicit assumption of 58–129 % for the LGM to present changes in SAT. Further LGM sensitivity simulations with isotope-enabled AGCMs would be necessary to constrain this uncertainty.
[1] Kino et al., J. Geophys. Res., 126(23). https://doi.org/10.1029/2021jd035397, 2021.
[2] Masson-Delmotte et al., Quat. Sci. Rev., 29(1), 113–128, https://doi.org/10.1016/j.quascirev.2009.09.030, 2010.
[3] Okazaki & Yoshimura, J. Geophys. Res., 124(16), 8972–8993. https://doi.org/10.1029/2018jd029463, 2019.
[4] Paul et al., Clim. Past, 17(2), 805–824. https://doi.org/10.5194/cp-17-805-2021, 2021.
[5] Sherriff-Tadano et al., submitted.
[6] Turner et al., Geophys. Res. Lett., 46(6), 3502–3511. https://doi.org/10.1029/2018gl081517, 2019.