Quality Control of Cloud Derived Wind Vectors from Geostationary Meteorological Satellites with Its Application to Data Assimilation System

Cloud derived wind vectors can provide plenty of useful information for synoptic analysis and numerical weather prediction, because of their excellent spatial and temporal coverages. It is of great value to apply cloud derived wind vectors efficiently with the purpose of improving the initial fields and numerical forecasts. The quality control of cloud derived wind vectors from geostationary meteorological satellites has been one of the important and difficult problems to be solved in satellite data assimilation. It has a direct impact on the prediction level of numerical weather prediction model. On the basis of the statistical analysis of global cloud derived wind vectors for 14 months, the quality indicator threshold of the five channels of cloud derived wind vectors in tropics, north and south hemisphere extra-tropics in high, middle and low levels from five global geostationary meteorological satellites in business operation, is analyzed and calculated. The error characteristics of infrared, water vapor and visible channels in the space area of cloud derived wind vectors are investigated, and the corresponding QI thresholds are determined. On the basis of quality control and equal-distance thinning scheme, the vertical distribution characteristics of cloud derived wind vectors are analyzed. It shows that after quality control, the number of infrared channel accounts for the vast majority, water vapor channel follows, and visible light channel the least.Furthermore, the method of innovation vector and zero-order Bessel fitting function which is based on the theory of least square method are adopted together to partition background and observation error variances in the observation space, and thus estimate the observation error of cloud derived wind vectors and the quality control coefficient according to the statistical distribution of the innovation vectors. On this basis, in order to validate the assumption on uncorrelated observation errors required in three-dimensional variational method, observation errors are inflated to avoid the influence carried by correlated errors.Finally, three numerical simulation schemes are designed and the forecast improvement and impact of cloud derived wind vectors before and after quality control are analyzed. Results show that the cloud derived wind vectors after quality control can effectively reduce the error of analysis field at high level. And the global short-term and medium-term forecast ability can be improved obviously by assimilating cloud derived wind vectors after quality control. In particular, there is a clear improvement in forecast ability in terms of wind, geopotential height and temperature fields in tropics. The forecast improvement above high levels appears better than those of middle and low levels in northern and southern hemisphere extra-tropics.