Study Of Data Assimilation Methods Of Ensemble Square Root Filter Based On Lorenz-96

"The blending of the existing noisy observations irregularly distributed in space and time into numerical models based on the physical laws that govern atmospheric flows became known as model assimilation of the data or data assimilation". It has become more and more important in the meteorology and oceanography due to the large number of available observations and increased computation ability. In the recent years, a new data assimilation method called an ensemble square root filter (EnSRF) has become popular. The available results indicate that EnSRF owns the potential ability of becoming an operational data assimilation method. So studying the EnSRF theory has an important scientific and practical meaning.Like ensemble kalman filter(EnKF), its background error covariance from the EnSRF is flow-dependent, while a fixed background error covariance is used in the operational variational data assimilation scheme at present. But the EnSRF does not need the perturbed observations which the EnKF needs and uses explicit mathematical transformations to achieve the analysis-error statistics. As we know that adding random perturbations to the observations would result in an additional source of sampling error related to the estimation of the observation-error covariance and may reduce the accuracy of the analysis covariance estimate. Therefore, the EnSRF data assimilation scheme is adopted in our study. The EnSRF also has drawbacks, for example, filter divergence, unbalance between the analysis variables et al, so we must find some ways to overcome them, for example, covariance inflation and reduction of spurious long range correlation.We test the EnSRF on Lorenz-96, a 40-variable dynamical model which has been widely used in the investigation on different data assimilation schemes. In this study, we will quantify the impact of different observation errors, sampling errors, the number of observations and the ensemble size et al...