A while ago, we have developed the Local Ensemble Transform Kalman Filter (LETKF) data assimilation system with the Weather Research and Forecasting (WRF) model (Miyoshi and Kunii 2011). We have been using 60-km resolution WRF, which produced good analyses and forecasts of Typhoon Sinlaku (2008) as shown in the left panel of the above figure. Several papers were published or are currently under review using this 60-km system. However, it is difficult to fight with the desire for a higher resolution.
Very recently, we have developed a new two-way nested WRF-LETKF system. The LETKF system was modified so that it can consider multiple resolutions simultaneously. Localization scales can be specified at each grid point separately. The enhanced LETKF and 60/20-km two-way nested WRF model enabled to produce a higher-resolution analysis as shown in the right panel. We start to see filaments of spiral rain bands of the tropical cyclone (color shades showing mm/6h rain rate). The eye is clearer, and the deep structure is more evident. Resolving capability is essential for these smaller-scale phenomena.
It is simply exciting as a scientist to have this better resolving capability. Now that we can see more details, we are going to investigate even more to improve tropical cyclone analyses and forecasts.
