Study On The Global Ozone Satellite Data Assimilation Experiment Based On Ensemble Kalman Filter

Ozone, as an important greenhouse gas, has an important influence on the global climate change. Climate Models provide a good platform for us to study global ozone distribution. Data assimilation technique provides an effective method for improving the accuracy of model prediction. As the development of data assimilation, Ensemble Kalman Filter gets more and more attention and is studied a lot. Compared with the conventional data,the satellite data has a great advantage in terms of resolution, covering area, and space and time consistency. Therefore, it is a meaningful topic that assimilate ozone satellite data using Ensemble Kalman Filter method to improve the model initial condition, and then, improve the accuracy of ozone forecast.In order to promote the assimilation of ozone in China, this paper build the CESM-ENSRF assimilation system based on the theory of Ensemble Kalman Filter and community earth system model (CESM).Then ozone satellite data assimilation experiments were designed to analysis the assimilation results and prediction results.The main research results and conclusions of this paper are shown as follows:(1) This paper successfully transplanted and install the CESM mode, through CESM forecast field can be seen that the model can simulate the global ozone distribution from the ground to the above of stratosphere, the vertical distribution structure is consistent with the actual situation. The CESM model is a good platform in study of the global ozone changes in the distribution.(2) The CESM-ENSRF system has been built based on the Ensemble Kalman Filter method. This system uses a modular design concept, consists of 6 main modules:input and output module, perturbation module, statistical analysis module, the main assimilate module, universal library and some peripheral processing program.(3) Several key problems in the assimilation of Kalman Filter have been considered:initial perturbation, variance inflation, variance localization, and the number of ensemble member and relevant schemes have been brought in.The two variance inflation schemes are compared by assimilation experiment, and the experimental results show that the best results is obtained by using two schemes at the same time.(4) To analyze the effect of the assimilation of the ozone satellite data on the model prediction, the system was constructed for the assimilation of the ozone profile data of the microwave limb sounding (MLS).The results show that: CESM-ENSRF system realizes the preliminary effect, model initial field has been significantly improved after assimilation, and the improvement of the initial field indeed has the positive effect on the model forecast field.