基于气象雷达反演和云图外推法的临近期降雨预报方法研究Research of approaching rainfall forecast method based on weather radar inversion and cloud image extrapolation
邸苏闯,李卓蔓,刘玉,潘兴瑶,郑琪,任黎,李永坤,薛志春
摘要(Abstract):
开展精准化临近期降水预报是开展实时洪涝预测预警的基础。传统的临近期降水预报一般基于气候模式进行数值预报,计算时间较长、空间分辨率低,预报精度仅为15%~20%,难以满足洪涝灾害预测预警需求。为快速、高精度、高空间分辨率预报降水过程,本研究提出了一种基于气象雷达反演和云图外推法的临近期降雨预报方法。主要包括:(1)采用金字塔光流法(Pyramid Lucas Kanade Optical Flow,简称LK光流法)进行云的运动矢量计算,根据运动矢量计算云团的运移位置;(2)提出基于回波强度和梯度特征的云层类型自动识别方法;(3)利用回波反射率与降水强度间的关系进行临近期降水量的反演及预报;(4)对预报结果进行落区精度评估和量级精度评估。2019—2020年四场降水实验研究结果表明:该方法可以实现0~1 h近实时1km高空间分辨率预测,而且预测期内的降水落区预测精度较高,平均精度可达60%;对落区内雨量站的量级精度进行对比分析得出四场降水的偏差BIAS分别为0.23 mm、3.56 mm、-7.12 mm、-5.15 mm,均方根误差RMSE分别为5.09 mm、8.45 mm、21.23 mm、14.63 mm,四场降水的相对误差绝对值均值为37.21%。预测精度均随时间增加呈下降趋势,下降程度与降水系统特征密切相关。总体来说,该方法对大范围高强度的降水预测精度较高,可为实时洪涝过程模拟分析提供重要支撑。
关键词(KeyWords): 气象雷达;云图外推法;临近期降雨预报;精度分析
基金项目(Foundation): 北京市重点科技计划课题(Z201100008220005);; 国家自然科学基金面上项目(41771393);; 江苏省研究生科研与实践创新计划项目(SJCX_20_0157);; 中央高校基本科研业务费项目(B200203187)
作者(Author): 邸苏闯,李卓蔓,刘玉,潘兴瑶,郑琪,任黎,李永坤,薛志春
参考文献(References):
- [1] 王坚红,黄维,王群,等.雷达回波预测极端暴雨概率方法构建原理与应用研究[J].高原气象,2015,34(2):575-585.WANG Jianhong,HUANG Wei,WANG Qun,et al.Construction and application of extreme rainstorm probability prediction based on radar echo base data[J].Plateau Meteorology,2015,34(2):575-585.
- [2] 白洁,王洪庆,陶祖钰.GMS卫星红外云图强对流云团的识别与追踪[J].热带气象学报,1997,13(2):158-167.BAI Jie,WANG Hongqing,TAO Zuyu.Recognition and tracking of strong convective cloud clusters from gms satellite infrared cloud images[J].Journal of Tropical Meteorology,1997,13(2):158-167.
- [3] WANG Y,LEE O.Use of two-dimensional deformable mesh struc-tures for video coding.I.The synthesis problem:mesh-based function approximation and mapping[J].Circuits and Systems for Video Technology,IEEE Transactions on,1996,6(6):636-646.
- [4] GANGULY A R,BRAS R L.Distributed quantitative precipitation forecasting using information from radar and numerical weather prediction models[J].Journal of Hydrometeorology,2003,4(6):1168-1180.
- [5] GOCHIS D J ,DAVIS C A .Second symposium on quantitative precipitation forecasting and hydrology[J].Eos Transactions American Geophysical Union,2013,87(36):363-363.
- [6] 杨文宇.基于天气雷达的城市降雨特征及临近预报研究[D].北京:清华大学,2017.YANG Wenyu.Investigation of spatial-temporal characteristics of urban rainfall and development of a nowcasting model based on weather radar observations[D].Beijing:Tsinghua University,2017.
- [7] 蔡涛,李德华,朱洲,等.基于彩色图像序列的特征检测和跟踪[J].计算机工程,2005,31(8):12-13,48.CAI Tao,LI Dehua,ZHU Zhou,et al.Feature detection and tracking based on color image sequence[J].Computer Engineering,2005,31(8):12-13,48.
- [8] 刘玉.基于雷达及静止气象卫星的遥感降水临近预报模型及可预报性研究[D].北京:中国科学院,2014.LIU Yu.Study on the model and predictability of remote sensing precipitation proximity prediction based on radar and geostationary meteorological satellite[D].Beijing:Chinese Academy of Sciences,2014.
- [9] 何立富,周庆亮,谌芸,等.国家级强对流潜势预报业务进展与检验评估[J].气象,2011,37(7):777-784.HE Lifu,ZHOU Qingliang,CHEN Yun,et al.Introduction and examination of potential forecast for strong convective weather at national level[J].Meteorological Monthly,2011,37(7):777-784.
- [10] 胡胜,孙广凤,郑永光,等.临近预报系统(SWAN)产品特征及在2010年5月7日广州强对流过程中的应用[J].广东气象,2011,33(3):11-16.HU Sheng,SUN Guangfeng,ZHENG Yongguang,et al.Features of nowcasting system (SWAN) products and its application in the process of strong convection in Guangzhou on May 7,2010[J].Guangdong Meteorology,2011,33(3):11-16.
- [11] 卢江涛.利用天气雷达估测和预报降雨分布的研究[D].北京:清华大学,2011.LU Jiangtao.Study on precipitation estimation and nowcasting based on weather radar[D].Beijing:Tsinghua University,2011.