高级检索
王洪, 杨洁帆, 龚佃利, 等. 2023. 基于微雨雷达的降水参数反演和粒子相态识别[J]. 大气科学, 47(3): 739−755. doi: 10.3878/j.issn.1006-9895.2201.21210
引用本文: 王洪, 杨洁帆, 龚佃利, 等. 2023. 基于微雨雷达的降水参数反演和粒子相态识别[J]. 大气科学, 47(3): 739−755. doi: 10.3878/j.issn.1006-9895.2201.21210
WANG Hong, YANG Jiefan, GONG Dianli, et al. 2023. Inversion of Precipitation Parameters and Precipitation Type Classification Based on Micro Rain Radar [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 47(3): 739−755. doi: 10.3878/j.issn.1006-9895.2201.21210
Citation: WANG Hong, YANG Jiefan, GONG Dianli, et al. 2023. Inversion of Precipitation Parameters and Precipitation Type Classification Based on Micro Rain Radar [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 47(3): 739−755. doi: 10.3878/j.issn.1006-9895.2201.21210

基于微雨雷达的降水参数反演和粒子相态识别

Inversion of Precipitation Parameters and Precipitation Type Classification Based on Micro Rain Radar

  • 摘要: 本文基于微雨雷达原始的后向散射信号,采用一种新的功率谱处理算法(RaProM算法),在功率谱计算、噪声去除、退模糊等处理的基础上计算了雷达基本参量, 并反演了液态降水参数,例如雷达反射率因子、雨强等, 并对粒子相态进行识别。RaProM算法综合考虑粒子下落速度、等效雷达反射率因子、不同相态粒子的尺度特征以及是零度层亮带位置等信息,可识别的粒子相态包括雪、毛毛雨、雨、冰雹以及混合相态。选取了三个山东地区较为典型的个例对RaProM反演算法进行验证,即2021年7月2日典型层状云降水个例、2019年12月25日雨雪转换个例以及2018年3月4日零度层高度逐渐降低的降水个例。结果显示:粒子识别方法应用于典型层状云降水,垂直方向上不同相态粒子的分层较为明显,过冷层里的固态降水雪花、零度层附近冰水转换区的混合相态降水以及零度层以下的液态降水符合现有认识,验证了反演算法以及粒子识别算法的有效性。将结果进一步在雨雪转换降水相态识别中和零度层高度的检测,该反演算法均能得到较好应用,与同址同步观测的微波辐射计、云雷达、二维视频雨滴谱仪等观测结论一致。另外,与微雨雷达标准反演算法对比,RaProM算法的优势是没有粒子相态的原始假设,且考虑降水粒子向上的速度,反演结果与微波辐射计、云雷达在垂直结构上有较高的一致性。与地面激光雨滴谱仪观测数据对比显示,也有效提升了微雨雷达对雨滴谱和雨强的反演能力。

     

    Abstract: Based on the original backscattering signals of the micro rain radar and a new micro rain radar processing methodology (RaProM), the equivalent radar reflectivity, the particle falling velocity, and Doppler spectrum width are calculated after power spectrum calculation, noise removal, and dealiasing. Furthermore, the precipitation types are identified. The RaProM algorithm can identify particle phases, such as snow, drizzle, rain, hail, and their mixed types, considering particle falling velocity, equivalent radar reflectivity, particle size characteristics of different precipitation types, and presence of bright bands. In addition, liquid precipitation parameters, such as radar reflectivity factor and rain intensity, are computed. Subsequently, three typical cases of stratiform cloud precipitation on July 2, 2021, transition of rain and snow on December 25, 2019, and the precipitation with the height of the zero degree layer decreasing gradually on March 4, 2018, are selected to verify and discuss the results. The method of precipitation type classification is applied to typical stratiform precipitation, the vertical structure shows snowflakes in the supercooled water area, mixed-type precipitation in the ice-liquid conversion zone near the 0°C layer, and liquid precipitation below the bright band, proving the validity of the method. The methods are then applied to precipitation type classification and bright band detection. The results show that the RaProM algorithm has the advantage of making no assumptions about precipitation type and considering particle upward velocity (such as snowflakes) over the standard inversion process of micro rain radar. The RaProM algorithm results are in good agreement with the colocated microwave radiometer and cloud radar in the vertical structure, and the deviations from the ground disdrometer in raindrop size distribution and rain intensity are reduced compared with the products of micro rain radar.

     

/

返回文章
返回