14:45 〜 15:00
[AAS05-05] Thermodynamics-Convection Coupling and Precipitation Characteristics In Global Km-scale Simulations
キーワード：全球嵐解像モデル、湿潤対流
Recent increase in computing power enables us to achieve a multi-year global simulation at kilometer (km)-scale horizontal resolutions. This storm-resolving earth system modeling partly helps reduce uncertainties in the representation of moist convection and clouds that can influence the climate system. Meanwhile, it is expected that there remain some biases originating from unresolved and/or underresolved processes at km-scale resolutions. To understand pros and cons of the storm-resolving framework, we compare the characteristics of precipitation and convection in multi-year simulations from three global km-scale models: ICON, IFS, and NICAM. We find a common issue of the underestimation of the convective cluster size, but the degree of this bias depends on the models. This diversity can be interpreted by differences in the representation of convection triggering and tropospheric moistening associated with deep convection. In particular, the former point is related to the convective sensitivity to lower-tropospheric moisture, affected by the treatment of turbulent mixing. Consistent with these convection characteristics, there are several notable differences in simulated specific weather/climate phenomena such as precipitation diurnal cycles (PDCs) and the Madden–Julian oscillation (MJO). The amplitudes and phases of PDCs over land are much more incongruous than those over ocean among the models. As for the MJO, IFS and NICAM reasonably simulate the eastward propagation of the large-scale convective envelopes, whereas ICON struggles to do so. These results highlight the importance of the continuous development and tuning of cloud microphysics and turbulent mixing in the free troposphere as well as the boundary layer at storm-resolving resolutions.