利用毫米波雷达Doppler功率谱和偏振参量反演云滴/冰晶谱分布及垂直气流的方法研究

Cloud droplet/ice crystal size distribution and vertical air velocity are the key parameters to describe the microphysical and dynamic process during the evolution of clouds. Currently, detection methods for such parameters are quite limited, while the Doppler power spectrum of vertically pointed millimeter-wavelength radar is an effective way. Due to the small terminal velocity of cloud droplets, the influence of vertical air motion upon cloud droplets is much greater than raindrops. How to remove the effect of radial wind on Doppler power spectrum becomes a main difficulty in retrieval. Referring to the previous studies and combining the polarimetric parameters of radar, this research project will develop a new retrieval technique which is suitable to retrieve the particle size distribution of cloud droplet, drizzle droplet, ice crystal with certain shapes and vertical air motion. In this technique, there is no need to make any apriori assumption of the types of particle size distribution, the phase of particles or vertical air velocity. Thus this technique has much wider application scope and better expansibility as compared with the existing international and domestic methods. Since the effect of vertical air motion has been fully considered, this technique is more suitable for cloud droplet with small terminal velocity. A fast inverse algorithm developed by this applicant will be used to solve this nonlinear inverse problem to ensure the retrieval accuracy and speed. The retrieval accuracy and the extent of application of this technique will be estimated by numerical simulations and field tests.

云滴/冰晶谱和垂直气流是研究云在生消演变中微物理和动力过程的重要参量。目前探测手段不多,垂直指向的毫米波雷达Doppler功率谱是一个新的反演云滴/冰晶谱和垂直风速的有效途径。云滴下落末速小,因此受垂直气流的影响远大于雨滴,如何去除径向风的影响一直是反演的难点。本课题拟借鉴国际上已有的反演方法，联合雷达的偏振参量，建立一个新的适用于Rayleigh近似条件下各种尺度的云滴、毛毛雨滴、特定形状冰晶的谱分布以及垂直气流的反演方法。与国内外已有方法相比，该方法不受粒子谱谱型和粒子相态的限制，无需对垂直气流作先验的假设，具有更为广泛的适用范围和更好的拓展性。由于充分考虑了垂直气流的影响，该方法更加适合反演下落末速较小的云滴。为保证反演速度和精度，拟采用申请者已建立的一套快速优化反演算法进行求解。通过数值模拟试验以及对实测功率谱的反演，评估方法的精度和适用范围。

摘要：滴谱（冰晶谱）是云在生消演变过程中形成的，是研究云中微物理过程和动力过程的重要参量。目前，利用垂直指向的毫米波雷达Doppler功率谱反演滴谱是研究云内微物理结构及动力过程的重要手段之一。本项目的研究内容主要包含以下几个部分：在建立毫米波雷达功率谱反演滴谱算法的基础上，利用数值模拟的方法详细分析了Mie散射效应、垂直气流速度对雷达反演诸要素（例如数浓度、雨强以及液态含水量）的影响，并将毫米波雷达反演滴谱与同期THIES滴谱仪观测数据进行了对比分析。结果表明，雷达反演的Z、I、N中值体积直径D0等在时间序列上与THIES滴谱仪都有较好的吻和度，其平均偏差分别为1.19DBZ、0.34mmh-1和0.36mm。这些结果均验证了雷达功率谱密度反演滴谱方法的可靠性。此外，结合不同类型冰雪晶与液滴粒子下落末速度的差异以及对云内最大下落末速度梯度（GFV）分析，本项目建立了一套针对层状云0℃层亮带高度及厚度的实时判别方法。利用该方法对山东省济南市一次层状云降水的分析表明，降水前20min 0℃层上方主要为霰，150min后0℃层上方主要是霰、枝状冰晶及液滴的混合物，这也显示了雷达功率谱反演冷云微物理结构的能力。