天气雷达定量估测降水量不同方法效果评估

Evaluation of the Precision of Weather RadarRainfall Estimation Algorithms

将黄河淮河洪水暴雨监测预报系统降水估测模式得出的规则网格点上的降水与雨量计实测值进行比较,结果表明:将地面雨量计值作为真值,则联合校准法和最优插值校准法得出的降水量计算精度最高,变分法得到的结果不是很理想,卡尔曼滤波校准法和平均校准法的计算精度低于联合校准法和最优插值校准法,Z-I关系法的精度最低。

Precipitation on ordered grid point,which is obtained from the Radar Estimating Precipitation model (REP) of Huanghe River and Huaihe River,is compared with the in situ data of a raingauge network.Results show that if the value of ground raingauge is taken as real value,then the optimal interpolation and joint calibration methods have the least root mean square errors,while the results of variational method are not very well due to parameter choosing.The calculating precision of Kalman filter method and average method is lower than the joint calibration method and optimal interpolation method.The method of Z-I has the largest root mean square errors.