[A-AS07] Weather, Climate, and Environmental Science Studies using High-Performance Computing

[AAS07-12] Comparison of stable water isotopes between the cloud-microphysical processes simulated by isotope-incorporated NICAM

*Masahiro TANOUE¹, Yuki Takano², Kei Yoshimura³, Hisashi Yashiro¹ (1.National Institute for Environmental Studies, 2.Atmosphere and Ocean Research Institute, The University of Tokyo, 3.Institute of Industrial Science, the University of Tokyo)

Keywords:NICAM, Stable water isotopes, cloud microphysical process

Stable water isotopes (SWIs) (δ²H and δ¹⁸O) show large spatial-temporal variability throughout the cloud process and large-scale transportation of atmospheric water vapor. However, there is some room for discussion of the effect of cloud-microphysics and large-scale transportation of water vapor on the variability of SWIs. Here, we incorporated SWI tracer into the Nonhydrostatic Icosahedral Atmospheric Model (NICAM) and investigated the contribution of cloud-microphysical processes to the variability of isotope ratios in precipitation and vapor. NICAM’s physical process can cover from low spatial resolution to high spatial resolution, and we can conduct simulations with two types of experimental setting; cloud-system-resolving-mode and general-circulation-mode (CRM-mode and GCM-mode). In the CRM-mode simulations, we used the single-moment bulk cloud microphysics scheme with six water categories (Tomita, 2008) for the cloud process, but did not use any convective parameterization scheme. In the GCM-mode simulations, we used the Arakawa-Schubert-type convective parameterization scheme and the large-scale condensation scheme as the cloud process with two water categories. We conducted an AMIP-type climate experiment from 1979 to 2000 using both settings with about 223 km of horizontal mesh resolution and 78 vertical layers, using 40 cores of the supercomputer Fugaku. The simulated δ¹⁸O values of precipitation and water vapor showed the latitude effect pattern (high δ¹⁸O in low latitude region, low δ¹⁸O in high latitude region), but those values in the CRM-mode were slightly lower than those in the GCM-mode. The simulated precipitation δ¹⁸O in the CRM-mode was lower in high altitude or inland regions compared with those in the GCM-mode. Besides, the precipitation d-excess, defined by δ²H – 8*δ¹⁸O, in the CRM-mode shows large spatial variability compared with the GCM-mode. These results indicates that the CRM mode show more larger spatial-temporal variability, meaning that the cloud-microphysical process is important for the large spatial-temporal variability of SWI. Although this study set the low spatial resolution, we will conduct these simulations with finer spatial resolution and a more extended simulation period.