10:45 〜 11:00
[AAS07-07] Preliminary study on model validation and data assimilation using EarthCARE
キーワード：EarthCARE、データ同化、モデル検証
A numerical weather prediction (NWP) model based on non-hydrostatic equations, incorporating a microphysics scheme, can predict the vertical motion of the atmosphere within cloud precipitation systems. However, the microphysics scheme contains errors due to the presence of unresolved processes and simplifications of processes for implementation. These errors contribute to biases in the initial conditions and forecast errors in the NWP system.
To identify the sources of these errors, we use EarthCARE satellite CPR (Cloud Profiling Radar) observation data (reflectivity and Doppler velocity) to validate the model predictions. When focusing on the detailed structure of clouds, the model with a horizontal grid spacing of 5 km shows low reproducibility, while the model with a grid spacing of 500 m shows improved reproducibility. Furthermore, we compare the detailed cloud internal structure predicted by large-eddy simulation models with EarthCARE observation data to investigate the issues within the current microphysics schemes.
Additionally, the impact of assimilating EarthCARE observation data into the initial conditions on forecast accuracy is investigated. In data assimilation, due to the limitations of linear approximation and computational resource constraints, low-resolution models are typically used. Therefore, a preliminary study is carried out to assess the effect of CPR assimilation on the forecast of large-scale meteorological phenomena, such as fronts.
In this presentation, an overview of these preliminary studies will be introduced.
To identify the sources of these errors, we use EarthCARE satellite CPR (Cloud Profiling Radar) observation data (reflectivity and Doppler velocity) to validate the model predictions. When focusing on the detailed structure of clouds, the model with a horizontal grid spacing of 5 km shows low reproducibility, while the model with a grid spacing of 500 m shows improved reproducibility. Furthermore, we compare the detailed cloud internal structure predicted by large-eddy simulation models with EarthCARE observation data to investigate the issues within the current microphysics schemes.
Additionally, the impact of assimilating EarthCARE observation data into the initial conditions on forecast accuracy is investigated. In data assimilation, due to the limitations of linear approximation and computational resource constraints, low-resolution models are typically used. Therefore, a preliminary study is carried out to assess the effect of CPR assimilation on the forecast of large-scale meteorological phenomena, such as fronts.
In this presentation, an overview of these preliminary studies will be introduced.