Research On Radial Velocity Quality Control And Circular Assimilation Of Multi-band Rada

Doppler radar is the main method for atmospheric remote sensing,which can not only be used for short-term near-term forecast analysis,but also for data assimilation to improve the forecast accuracy of numerical models.However,the radial velocity of the radar often suffers from velocity ambiguity,which affects its analysis and application.Moreover,assimilating radar data can also introduce erroneous observations into the model,which affects the improvement of the forecast accuracy.In this study,we conducted quality control and cycle assimilation of radar radial velocity data.We selected the UNRAVEL dealiasing algorithm to perform quality control and error analysis of the radial velocity base data.Based on this,we combined the GSI-3DVar assimilation system and the WRF numerical model to perform cycle assimilation of radar data.We investigated the impact of radial velocity quality control and cycle assimilation on the simulation of a severe thunderstorm gale in Nantong on April 30,2021,and drew the following conclusions:(1)After noise elimination,velocity dealiasing,and error analysis,the radial velocity of the radar effectively eliminates noise and velocity ambiguity,improving the accuracy of the data.Combining with quality control and effect evaluation of typhoons,thunderstorm gale,and heavy convective weather processes,the results show that the UNRAVEL algorithm can effectively eliminate velocity ambiguity of different convective weather and different band radar radial velocities after quality control.(2)By comparing the combined reflectivity and vertical profiles observed and simulated using several schemes,it was found that the use of radar radial velocity quality control and cycling assimilation effectively improved the initial field and simulation results of the model,including features such as merging of convective cells,bow echoes,and convective line splitting.The vertical profiles also better simulated the characteristics of strong echo vertical distribution and forward tilt of the centroid descent.By comparing the 10 m wind field observed and simulated using several schemes,it was found that after radar radial velocity quality control and cycling assimilation,the simulated results were closer to the ground truth observations,mainly reflected in the trend of wind speed evolution,the timing of the maximum wind speed value and its change,which were consistent with the observations.Therefore,the cycling assimilation after radial velocity quality control had the best improvement effect on the wind field simulation.(3)From the perspective of thermodynamic fields,the improvement of dynamic and thermal structures in the simulation after radial velocity quality control and cycling assimilation was more significant.The upper-level divergence and lower-level convergence form a "cold-warm-cold" thermodynamic structure,accompanied by strong upward motion,a north-high and south-low pressure distribution,and strong vertical wind shear,which helps to transfer the horizontal momentum of the mid-upper level to the near-surface layer through downdrafts,thereby triggering this catastrophic thunderstorm gale.