GPM卫星和地面雷达对江苏盐城龙卷风强降水估测的对比分析研究
投稿时间:2018-10-27  修订日期:2019-01-25  点此下载全文
引用本文:
摘要点击次数: 67
全文下载次数: 0
作者单位E-mail
黄朝盈 中山大学 huang.chaoying@163.com 
张阿思 中山大学大气科学学院  
胡宝清 南宁师范大学北部湾环境演变与资源利用教育部重点实验室  
陈生 中山大学 大气科学学院 chensheng@sysu.edu.cn 
基金项目:广西高校高层次人才培养与教师技能提升计划(8844),中山大学“百人计划”急需青年杰出人才项目(74110-52601108),四川大学水力学与山区河流开发保护国家重点实验室开放基金(No. SKHL1501),国家自然科学基金(51579162,51879174)
中文摘要:为综合评估卫星和天气雷达在2016年6月23日盐城龙卷风期间的强降水过程的降水估测精度,以国家级雨量站观测数据为基准,结合相关系数(CC)、相对误差(RB)、均方根误差(RMSE)以及分级评分指标,利用S波段的天气雷达定量降雨估测(RQPE)和全球降水观测计划多卫星融合产品(IMERG_FRCal,IMERG_FRUncal,IMERG_ERCal)进行比较。结果表明:研究区内,雷达和卫星的累积降水量与雨量站的空间相关性很强(CC>0.9),基本上能捕捉到整个降水过程的空间分布。其中雷达与雨量站的相关性最强(CC = 0.97),降水主要分布在江苏中部及北部,但卫星不能捕捉到江苏北部的强降水区域;对于小时时序区域平均降水,卫星和雷达均显著低估了累积降水量,低估值在-5.26%到-18.84%之间。综合降水中心区域分析,IMERG的RB值上升至-64.07%~66.81%,表明IMERG的强降水区域降水量与雨量站的时间序列的偏差显著;RQPE在降水峰值达到之前与峰值之后与地面雨量站的变化趋势基本一致,但对降雨量峰值(07~12时)有明显的低估。RQPE能较为准确地在时间上捕捉到降雨强度的变化趋势,但对于大雨及暴雨的估测能力不佳;RQPE的POD、CSI、FAR值都远远高于IMERG,IMERG几乎未能监测到强降水的发生,而RQPE在40~45mm/h区间存在最低值。总体上,RQPE对此次龙卷风强降水量的估测表现优于3种IMERG产品,特别是在捕捉强降水区域的空间分布明显表现比卫星产品佳,但对于强降水的估测能力仍需进一步改善。
中文关键词:盐城龙卷风,天气雷达,IMERG,定量降水估测 ,精度分析
 
Inter-comparison of Rainfall Estimates from Radar, Satellite and Gauge During Yancheng Tornado
Abstract:The study evaluates rainfall estimates from ground radar network against Early and Final run Integrated Multi-Satellite Retrievals for Global Precipitation Mission (IMERG) during an extreme precipitation storm over Yancheng Tornado on June 23, 2016. Statistics indexes used in this study included: correlative coefficient (CC), relative bias (RB) and root mean square error (RMSE), probability of detection (POD), critical success index (CSI) and false alarm ratio (FAR). The results shows that: 1) radar quantitative precipitation estimates(RQPE) and three IMERG all can capture the spatial pattern of storm cumulative rainfall with high CC from 0.90 to 0.97, and RQPE was well correlated with gauge measurement (CC ~0.97); 2)the IMERG can’t capture the rainfall center in northern Jaiangsu province; 3) both satellite and radar significantly underestimated the hourly area-mean cumulative precipitation from -5.26% to - 18.84%,the RB of IMERG varies from -64.07% ~ -66.81%, indicating the significantly deviation between the time series of IMERG and observations in the heavy rain storm; 4) RQPE agrees well with the trend of gauge observations before and after the peaks of rainfall, but underestimates the maximum rainfall; 5) the RQPE can capture well the trend of rainfall intensity in terms of space and time, but performs less well in estimating heavy rain and torrential rain (>5mm/h); those of 6)the POD, CSI and FAR values of RQPE are much higher than IMERG. IMERG hardly detect the occurrence of heavy precipitation(>25mm/h), while RQPE has the lowest value in the range of 40~45mm/h. Overall, radar outperforms satellite in estimating precipitation during extreme rainfall storms, but still needs to be improved to capture the intensive rainfall peaks.
keywords:Tornado  GPM  Weather radar  Quantitative precipitation estimation
  查看/发表评论  下载PDF阅读器