Polarimetric Radar and Surface Data Assimilation Using a Nested-LETKF System and an Ensemble-based Sensitivity Analysis for the Tornadic Supercell on 6 May 2012

A band-shaped precipitation system spawned almost simultaneously three tornadoes 50-90 km northeast of Tokyo in the Kanto Plain at about 1230 LT on 6 May 2012. The southernmost tornado is one of the strongest tornadoes in Japan (estimated to be F3 in the classical Fujita scale). A low-level mesocyclone (LMC) associated with this tornado was well observed by Meteorological Research Institute advanced C-band solid-state polarimetric radar (MACS-POL), which is located within about 15 km distance. Using a nested Local Ensemble Transform Kalman Filter (LETKF) system with 32 ensemble members through assimilation of dense surface meteorological data (horizontal wind, temperature and relative humidity) observed by Automated Meteorological Data Acquisition System (AMeDAS) and Environmental Sensor Network (ESN), radial wind observed by four C-band radars, and rainwater estimated from reflectivity and specific differential phase observed by MACS-POL, we have successfully forecasted the tornadic supercell associated with this LMC. Furthermore, the factors controlling the strength of the forecasted LMC are clarified by the ensemble-based sensitivity analysis (ESA). The ESA shows that the strength of the LMC is sensitive to low-level southerly winds in the front side of the storm and low-level water vapor in the rear side of the storm 1-hour before the LMC genesis. If these low-level variables are not modified by the assimilation of either surface meteorological data or radar data in the LETKF analysis, the strong LMC is not forecasted near the path of the actual tornado. This demonstrates the modification of these low-level variables is important for forecasting the LMC in this case.