基于Ka/Ku双波段回波强度差约束和多普勒功率谱的微物理和动力参数反演方法和应用

Air Vertical Motion and Raindrop Size Distribution Retrieval Algorithm Based on Reflectivity Spectral Density Data and Dual-Wavelength Ratio Constraint with Ka/Ku Dual-Wavelength Cloud Radar and Its Preliminary Applications

回波强度定标误差、天线水膜衰减和雨区衰减造成的回波强度偏差对云雷达反演微物理和动力参数有非常重要的影响,准确分析这些偏差对提高反演精度至关重要。为了消除云雷达因定标和天线罩等引起的回波强度和功率谱大小的影响,实现高精度和雷达全观测范围的反演,本文提出了基于Ka/Ku双波段云雷达回波强度差约束和回波强度谱密度数据的降水内空气垂直运动速度和雨滴谱反演方法(DWR-SZ),并将该方法应用到2020年6月8日和2021年6月1日华南两次对流性云降水垂直结构观测数据,利用雨滴谱仪数据分析了该方法反演结果的改进程度,分析了上升速度对反演的回波强度和微物理参数的影响。该方法首先融合双波段云雷达反演(DWSZ)和单波段小粒子跟踪方法(ST)方法反演的云内空气垂直速度V_(air),形成全观测域的V_(air),然后利用DWSZ方法得到微物理参数初估值,并计算衰减影响,最后利用双波段回波强度差(DWR)调整回波强度系统偏差和反演的微物理参数,使DWR-SZ方法正演得到的DWR与雷达观测值差到达极小。结果表明:(1)采用脉冲压缩技术的高雷达灵敏度模式与采用短脉冲的低灵敏度模式相比,DWSZ方法反演的V_(air)与雷达灵敏度相关性非常小,结果稳定,但这种方法只能应用于含有大粒子的液体降水区(粒子直径大于1.8mm);小粒子跟踪ST方法通常低估V_(air),但在低层的35 dBZ以下降水V_(air)低估程度不大,且灵敏度提高会极大改进Ka波段雷达反演能力;两种方法融合的V_(air)比较合理;(2)雨区衰减和距离是造成ST方法低估V_(air)的主要原因;而固态降水的功率谱非常窄而且陡,灵敏度对固态降水区V_(air)影响不大;(3)采用DWR作为约束,有效减小了回波强度的系统偏差和天线水膜影响,提高了微物理参数的反演准确率;(4)ST方法反演的V_(air)高估了粒子数密度,液体含水量(LWC)和衰减系数,低估了粒子大小,但对天线水膜引起的回波强度系统偏差影响不大。

Reflectivity calibration errors,attenuation in regions prone to rainfall,and water cover over cloud radar antennas have severely impacted the microphysical and dynamic parameters retrieved using reflectivity spectral density data;examining the errors involved with retrieving microphysical and dynamic parameters is a crucial problem that needs resolution.This paper presents a retrieval algorithm for vertical air motion(V_(air))and raindrop size distribution(DSD)based on reflectivity spectral density data and dual-wavelength ratio(DWR)constraints with a Ka/Ku dual-wavelength cloud radar(DWCR),which aims to reduce the effects of observation bias introduced by calibration and attenuation of water cover over cloud radar antenna on reflectivity.The disdrometer data were employed to analyze the retrieved parameters.Furthermore,the effects of vertical air speed on retrieved microphysical parameters are discussed.In the algorithm(DWR-SZ),V_(air)retrieved from a single Ka/Ku band CR(ST)and DWCR algorithms(DWSZ)are merged to form V_(air)in all observation areas,following which the initial DSD and attenuation are retrieved using the DWSZ algorithm.Finally,DWR between the first and last ranges in the liquid area in a beam is utilized to adjust the reflectivity bias and the retrieved final DSD to minimize the difference between the observed cloud radar and the calculated DWR.Two convective precipitation cases on June 8,2020,and June 1,2021,in Longmen,Guangdong Province,were used to examine the retrieved results.The results demonstrate that the radar sensitivity variations have little effect on V_(air)obtained from the DWSZ;however,the DWSZ cloud is only employed in the areas containing large raindrops(diameter>1.8 mm).ST algorithms with Ka and Ku data underestimated V_(air);however,the V_(air)r is reasonable at low altitudess with reflectivity weaker than 35 dBZ.A highly sensitive work mode with pulse compressions could enhance the V_(air)retrieval bias.Merged V_(air)from the ST and DWSZ algorithms is reasonable.The attenuation and far radar range could introduce the underestimation of V_(air)with ST algorithms.Moreover,the underestimations of V_(air)in the solid precipitation area are negotiable due to the sharp variation and narrow reflectivity spectral density data.Furthermore,employing the constraint conditions of DWR reduced the bias of observed reflectivity and the effects of water cover over the antenna,thereby improving the retrieval results.V_(air)from the ST used in DWR-SZ overestimated the drop number,liquid water content(LWC),and attenuation coefficient and underestimated the drop size;however,it has no effect on the reflectivity bias produced by water cover over the antenna.