[AAS01-P01] Ultra-high Resolution Numerical Weather Simulation and Dependency of Simulated Convective Cells on Model Resolutions
Keywords:heavy rainfall simulation, high-resolution model, K computer
This study investigates the impact of a numerical weather prediction model's factors (horizontal resolution and planetary boundary layer schemes (PBL)) on heavy rainfall simulation. The case study is the Hiroshima heavy rain event in August 2014. The model resolutions are 5 km, 2km, 500 m, and 250 m, and the PBL schemes are the Mellor-Yamada-Nakanishi-Niino level3 and the Deardorff scheme.
The higher resolution (500-m and 250-m grid spacing) models reproduced a more accurate location and intensity of the rainband than the lower resolution (5-km and 2-km grid spacing) models. The PBL schemes had a smaller impact on this case.
This study also investigates the dependency of simulated convective cores (CCs) on model resolutions. The local rate of change of the number of CCs with respect to the model resolution has been found to start decreasing at very high resolutions which are around 500-m grid spacing. This implies the number of CCs tends to converge when the resolution goes higher beyond 500 m. In summary, this study has demonstrated the benefit of using a high-resolution model (500-m grid spacing or less).
The higher resolution (500-m and 250-m grid spacing) models reproduced a more accurate location and intensity of the rainband than the lower resolution (5-km and 2-km grid spacing) models. The PBL schemes had a smaller impact on this case.
This study also investigates the dependency of simulated convective cores (CCs) on model resolutions. The local rate of change of the number of CCs with respect to the model resolution has been found to start decreasing at very high resolutions which are around 500-m grid spacing. This implies the number of CCs tends to converge when the resolution goes higher beyond 500 m. In summary, this study has demonstrated the benefit of using a high-resolution model (500-m grid spacing or less).