Design Of Surface Data Assimilation Approach

Content: Surface observation data are widely distributed and of high density. Besides, the observation elements are directly used in numerical models. How to use the information of surface observation data effectively to improve the accuracy rate of mesoscale numerical weather prediction is a meaningful research. Elevation difference between surface observation sites and numerical model surface is one big problem in SODA (Surface Observation Data Assimilation). On this issue, a heavy rainfall during 5th-7th June 2006 in Fujian province is simulated using the WRF mesoscale numerical model firstly. Then some work have been done to SODA with WRF₃DVAR system. On the base of Ruggiero surface data assimilation approach, a new surface observation data assimilation method adapted to topography in our country is put forward to solve the influence of different elevation between observation stations and model surface. Results presented in this thesis can be summarized as follows: (1) It is a typical South China quasi-stationary precipitation that is based on stationary Front. With the help of Bell-mouthed terrain and weather-side of Wuyi Mountain, southwest airflow veered and lead to the convergence at low level which resulted in the form of mesoscale shear and vortex. Warm and wet air was raised by the mountain, then enhanced the severe convection at middle-level, all above reinforced the rainfall on the windward slope.(2) Compared with Ruggiero assimilation approach, surface data is utilized more sufficiently in new approach without elimination while better analysis and simulation results can be obtained. (3) Because elevation of surface observation stations is taken into account in the new approach, the temperature increment corresponds with the elevation difference between surface observation station measurement and numerical model surface. (4) Temperature and humidity assimilation can impact the analysis field from bottom to top of numerical model, and brings more contribution to the mid-lower layers' analysis field while less to higher layers, and the impact of temperature is more distinct. The precipitation simulation is sensitive to initial temperature field.