| dc.description.abstract | This study uses the Weather Research and Forecasting (WRF) model and three-dimensional data assimilation (3DVAR) system to ingest various observations, including FORMOSAT-3 GPS RO, DOTSTAR dropsondes, SSM/I, QuikSCAT, and conventional soundings, into the WRF model for understanding the impact of these data on improvement of numerical weather prediction. Five cases of 2008 weather systems, including two Mei-yu cases and three typhoon cases, are selected for simulation.
The model initial increments show that assimilation of SSM/I or GTS gives humidity increments larger than those from the experiments assimilating other observations. Temperature increments are, however, larger for assimilation of GPS or GTS than other observations. Assimilation of dropsondes or QuikSCAT obtains the most significant wind increments. The results of Mei-yu and typhoon simulations show that assimilation of GTS or SSM/I data simulation has the best improvement, while QuikSCAT data also have some positive impact on typhoon forecasts. Assimilation of GPS RO data does not have a clear impact on typhoon track forecast. On the other hand, assimilation of dropsonde soundings tends to result in a faster moving system. For rainfall prediction, assimilation of GPS or SSM/I appears to give the best performance. The control experiment without assimilation of any observational data in general gives the worset rainfall prediction. The results indicate that consistent modifications on water vapor mixing ratio and wind fields are important to the weather forecasts.
Among the impacts of combined observational data, the dropsonde soundings appear to be more dominant when the GPS data are also assimilated. When GPS data are also assimilated, the simulation with assimilation of other data will become improved. In GPS cycling run experiments, significant improvement of prediction is found in simulations with assimilation of other observations. From the sensitivity tests, single GPS sounding in the vicinity of typhoon circulation might have a considerable impact on typhoon prediction.
| en_US |