利用雷达资料对自动雨量计实时质量控制的方法研究

自动雨量计资料是对降水的直接测量,在流域面雨量计算、气候研究、气象服务等方面具有重要意义。但是,由于风力、蒸发、灌溉、校准、漏斗堵塞、机械故障、信号传输等原因往往造成其存在不同类型的系统误差和随机误差, 自动雨量计数据在定量使用前需要进行质量控制。目前,天气雷达以其高时空分辨率的优势已经成为监测降水的重要手段,本文首先采用两步校准法改善雷达估测降水,然后对雷达—雨量计对之间的差异进行统计学的分析,确定自动雨量计质量控制的一些标准,从而对雨量计进行质量控制。最后用两个降水过程对自动雨量计质量控制的结果进行了检验,结果表明:两步校准法改善了雷达估测降水的系统性偏差,并减小了雨量计站点上的相对误差;可以利用雷达估测降水实现对自动雨量计的实时质量控制,就整个数据集而言,约0.1%的数据被怀疑为误判,误判的自动雨量计主要位于雨带的边缘。但该质量控制算法同时也存在一定的局限性:在雨带的边缘或没有天气雷达覆盖的区域,以及雷达资料存在数据质量问题的情况下,往往会造成对雨量计的误判。

Development of a Real-Time Quality Control Method for Automatic Rain Gauge Data Using Radar Quantitative Precipitation Estimation

Automatic rain gauges measure precipitation directly and are important in areal rainfall calculation, climate research, and meteorological services. However, a system for quality control is required when using automatic rain gauge data quantitatively due to various types of systematic and random errors caused by wind, evaporation, splashing, calibration, funnel blockage, mechanical failure, finite sampling, signal transmission interference, power failure, and other factors. Doppler radar has become an important method for monitoring precipitation in recent decades due to its high spatial and temporal resolution. In this study, a two-step calibration method was used to improve radar quantitative precipitation estimation. Then, the differences between radar-gauge pairs were statistically analysed to determine an effective criterion for rain gauge quality control. Finally, two heavy rain events were studied to assess the proposed quality control procedure. The results showed that the two-step calibration method can improve radar quantitative precipitation estimations by removing systematic bias and eliminating relative errors in space. Thus, radar quantitative precipitation estimation can be applied in real-time automatic rain gauge quality control. With respect to the entire set of gauge data, the percentage of false quality control results was as suspected, approximately 0.1%, with the wrongly rejected gauges located mainly on the edges of rain bands. The real-time rain gauge data quality control method, however, causes the incorrect rejection of gauges not only in regions where rain gauges are located on the edges of rain bands but also in regions where radar data are unavailable or are not under quality control.