Observing System Impact Test And Predictability

The atmosphere is a typical chaos system, in which there is an inner uncertainty, and the errors coming from the initial values and model of the numerical weather prediction always exist. Therefore, in many research topics the predictability of the atmospheric movement is an interesting problem. Recently, the study on the atmospheric prediction have profoundly and extensively developed. As to the typical problems, besides the new estimates in the theoretical limitations of the atmospheric predictability, more efforts have been done in the studies on the actual predictability instead of theoretical predictability, and the correct understand of the atmospheric predictability has b ecome t he i mportant b asis o f t he d evelopment o f t he n umerical weather prediction and the data assimilation. It is significant how to make use of the study conclusion of the prediction in the meso-scale and adequately to process data assimilation, ensemble prediction, adaptive observation and so on.In this paper, the main tool is the Global/Regional Assimilation PrEdiction System (GRAPES) which is exploited by the Chinese scientist. And the experiments have been carried out from the routine observation and the adaptive observation. Furthermore, the impacts of the routine observation system in China on the numerical weather prediction are estimated and the impacts in the typical cases of the TEMP data and the dropsonde data are studied. Additionally, the dynamics method is used to attempt to find the sensitive regions and the idea of the ensemble prediction is used to process the perturbation impact studies in the sensitive regions. The conclusions show that the strengthen of the studies on the impact studies of observation systems and the development of the adaptive observation can improve the forecast capabilities and skills in the extreame weather events in China. The following is the main summaries of this paper:1. The conclusions of the impact studies of uncertainty and the impact studies of routine observation system show: a) In the cases, the difference from the initial value is more important than one from the model difference between the hydrostatic balance and n on-hydrostatic b alance. b) T he T EMP d ata h as a n o bvious p ositive i mpact t o improve the forecast of the regional numerical weather prediction in China, and the SATOB data and SHIP data also help to improve the forecast of the region in China, but the impact of which are less than that of the TEMP data, c) However, the impacts of the TEMP data sometimes are negative which are often found in the forecast after 48 hours, d) Besides, the TEMP data are important to impact on the forecast of the temperature and the precipitation.2. The studies are carried out on the impact of a new type of adaptive or targeting observation, i.e. dropsonde observation, on the typhoon DUJUAN. And it shows: a) After adding the dropsonde data to the 3DVAR assimilation system, the forecasts of the elements have been improved obviously, such as the path, the center location, and the intensity of typhoon and the geopotential heights, but the forecast of the movement velocities oftyphoon becomes worse. b)The forecasts of the wind andprecipitation of typhoon are also improved with the dropsondes data, c) Furthermore, it was found that in the 3DVAR assimilation system the deviation structure impacted on the forecast largely.3. In view of the importance the TEMP data, a series of experiments are carried out to find the sensitive regions of the analysis field in the case, and the perturbation impact studies are performed in the sensitive regions. The conclusions are as following: a) The difference of the geopotential heights of the initial field in the sensitive regions spread to the downstream with the airflow and grow. At the same time, a new negative difference region occurs in the downstream. And the positive and negative difference regions both go round and round anticlockwisely with the airflow and grow in the low pressure cycle, b) The observations in the sensitive region improve the 48...