Short‐term forecasting through intermittent assimilation of data from Taiwan and mainland China coastal radars for Typhoon Meranti (2010) at landfall

Radial velocity (Vr) and reflectivity (Z) data from eight coastal operational radars of mainland China and Taiwan are assimilated for the first time using the ARPS 3DVAR and cloud analysis package for Pacific Typhoon Meranti of 2010. It is shown that the vortex‐scale circulations of Meranti can be adequately established after only 2 hourly assimilation cycles while additional cycles provide more details for subvortex‐scale structures. Subsequent 12 h forecasts of typhoon structure, intensity, track, and precipitation are greatly improved over the one without radar data assimilation. Vr data lead to a larger improvement to the intensity and track forecasts than Z data, while additional Z data further improve the precipitation forecast. Overall, assimilating both Vr and Z data from multiple radars gives the best forecasts. In that case, three local rainfall maxima related to typhoon circulations and their interactions with the complex terrain in the southeast China coastal region are also captured. Assimilating radar data at a lower 3 or 6 hourly frequency leads to a weaker typhoon with larger track forecast errors compared to hourly frequency. An attempt to assimilate additional best track minimum sea level pressure data is also made; it results in more accurate surface pressure analyses, but the benefit is mostly lost within the first hour of forecast. Assimilating data from a single Doppler radar with a good coverage of the typhoon inner core region is also quite effective, but it takes one more cycle to establish circulation analyses of similar quality. The forecasts using multiple radars are still the best.