Evaluation of a vortex relocation scheme and the impact of various synthetic observations on tropical cyclone initialization in an advanced research WRF

The accurate specification of the tropical cyclone (TC) inner-core structure has a great impact on its numerical forecast. In this study, a series of numerical simulations of two TCs (Typhoons Nuri and Hagupit 2008) over the Western Pacific ocean are conducted using the advanced research version of Weather Research and Forecasting (WRF ARW, V3.1) model and its 3-dimensional data assimilation system to evaluate a vortex relocation scheme and analyze the impact of assimilating different TC synthetic observations over the vortex inner-core regions.

A TC vortex relocation scheme is implemented into WRF ARW based on the GFDL storm relocation scheme. The basic field , disturbance field, analyzed hurricane vortex and the environmental filed are obtained during the TC relocation. The hurricane vortex is adjusted to the accurate position according to TC best track observations from the Joint Typhoon Warning Center (JTWC). The synthetic observations are then assimilated into the model in order to obtain a realistic vortex structure.The impact of each type of the synthetic observations (e.g.sea level pressure, wind, temperature and moisture over the vortex core areas) on the vortex initialization is then examined.

Results show that (1) the TC vortex relocation scheme results in improved TC track forecasts; (2) Compared with the use of surface data, the assimilation of observations at multiple vertical levels has a better effect on the TC initialization and forecasts; (3) Wind observations are more useful than temperature observations in some cases. However, the use of temperature and relative humidity data together results in better numerical simulations of TCs; (4) The direct assimilation of synthetic observations into the model leads to much better TC forecasts than directly implanting the bogus vortex into the model.