Data Quality Control For Weather Radar Reflectivity Factor And Its Application In Regional Radar Composite

Radar is currently the only observation system that may accurately describe the convective-scale precipitation phenomenon.Making use of Doppler radar data,extracting and utilizing meaningful meteorological information is of great value for forecasting and warning severe weather and is also the most important source of reference for Nowcasting.During the detection process,Doppler weather radar data is inevitably affected by beam blockage,electromagnetic wave attenuation and ground clutter which causes quality problems in radar data and affects the quantitative use of radar data.Therefore,it is of great scientific significance and practical significance to study various quality problems of radar data.How to identify and correct the relevant echoes and improve the data quality of radar is the key to this study.This paper mainly focuses on the identification and correction of beam blocking,the identification and removal of clutter and the attenuation of electromagnetic waves.The main contents are summarized as follows:(1)According to the spatial correlation of radar observation data,a beam blocking recognition and correction algorithm that does not depend on high-precision digital elevation model is proposed,which can be conveniently used for blocking recognition and correction of weather radar reflectivity factor.For inaccurate echo data caused by partial blockage or small-range complete blockage,blockage recognition and linear interpolation are performed to correct the block data.Three different types of precipitation data from NUIST-CDP radar are selected to verify the algorithm.Result shows that the algorithm can effectively correct the block data and improve the radar data quality.(2)According to the ground clutter that appeared in the radar data,the ground clutter echo is identified based on the radar echo texture feature algorithm,the standard template algorithm and the fuzzy C-means algorithm,and the ground recognition algorithm with the polarization parameters is used to evaluate the above three algorithms as the standard.Result shows that the terrain template algorithm is greatly affected by the seasons and climate,and the FCM clustering method is easy to misidentify the clear sky echo as the ground echo,resulting in the partial clear sky echoes missing.The texture feature method is convenient for calculation and best suited for professional use because of small error.(3)A new method,Radial Fitting,combining with the advantages of C and S-band radar systems,is proposed for attenuation correction of NUIST-CDP radar and compared with the revised result of the ZH-KDP method.The S-band radar observation data with smaller attenuation is spatially interpolated to the same spatial resolution as NUIST-CDP.Wavelet analysis method is used to filter radar ?Dp data first,and then variable distance least squares method is used to re-evaluate KDP.Comparing the profile of the radial echo intensity with distance before and after the correction,and analyzing the variation of the scatter distribution of the polarization parameter KDP?ZH,the result shows that the corrected echo intensity values of both methods are increased,ZH-KDP method has a small amount of correction at a short distance and a large amount of correction at a long distance,it is affected by the empirical coefficient and generates a certain amount of error.The results of the radial fitting method are in good agreement with the actual measured value in the S-band radar,it is not affected by the relationship between the reflectivity factor and the attenuation rate,the raindrop spectrum,and the radar data quality can be significantly improved.(4)This chapter selects S-band dual-polarization Doppler weather radar data from four stations in Qingyuan,Shaoguan,Yangjiang,and Guangzhou in Guangdong Province for regional network.The radar reflectivity factor regional network of typhoon "Nida" landing in Guangzhou is analyzed on the basis of quality control of radar data at the height of three kilometer.The analysis shows that multi-site radar data regional network can exploit the advantages of multi-radar detection and overcome the limitation of the maximum detection distance of a single radar,providing high spatial resolution data information for the analysis of large-scale weather systems.