Search

Initial condition is difficult to determine for Numerical Weather Prediction of rainfall prediction. Data assimilation (DA) is expected to improve the initial condition in the model. This study examines the influence of the low-temperature on the generation of convective clouds. A wider region is used for assimilation in to evaluate low-temperature area in middle atmosphere. A case study in west Japan is performed for verification. As a result, the convective clouds over Rokko mountains can be predicted due to reproduction of low-temperature mass by data assimilation.

The short lead time rainfall prediction by Numerical Weather Prediction model has some difficulties in the spin-up problem. Therefore, data assimilation (DA) is expected to improve the initial condition in the model. In this study, our developed ensemble DA system, CReSS-LETKF, and the method of estimation of ice-water mixing ratios are employed. DA of rain, graupel, ice crystal, snowflake and Doppler velocity estimated by polarimetric Doppler radar are carried out after the first convective cloud in mesoscale convective systems is generated. As a result, the first convective clouds formed in initial condition have effective influence on the short lead time rainfall prediction. As the next challenging step, DA is carried out before the first convective cloud. As a result, convective clouds are not generated although the atmosphere conditions, such as potential temperature change

Initial condition is difficult to determine for Numerical Weather Prediction of rainfall prediction. Data assimilation (DA) is expected to improve the initial condition in the model. This study examines the influence of the low-temperature on the generation of convective clouds. A wider region is used for assimilation in to evaluate low-temperature area in middle atmosphere. A case study in west Japan is performed for verification. As a result, the convective clouds over Rokko mountains can be predicted due to reproduction of low-temperature mass by data assimilation.

The short lead time rainfall prediction by Numerical Weather Prediction model has some difficulties in the spin-up problem. Therefore, data assimilation (DA) is expected to improve the initial condition in the model. In this study, our developed ensemble DA system, CReSS-LETKF, and the method of estimation of ice-water mixing ratios are employed. DA of rain, graupel, ice crystal, snowflake and Doppler velocity estimated by polarimetric Doppler radar are carried out after the first convective cloud in mesoscale convective systems is generated. As a result, the first convective clouds formed in initial condition have effective influence on the short lead time rainfall prediction. As the next challenging step, DA is carried out before the first convective cloud. As a result, convective clouds are not generated although the atmosphere conditions, such as potential temperature change