Abstract: Based on the data of wind profiler and radiosonde of Beijing observatory station from 2014 to 2018, and the surface data of MICAPS in the same period, the errors and correlations of horizontal wind between wind profiler data and sounding data were compared and analyzed under different precipitation conditions. It could provide reference for application of wind profiler data in precipitation forecast. The results are as follows: (1) In general, the U-component wind of wind profiler was smaller than sounding data at all levels. The error of the U-component wind between the wind profiler data and sounding data was small and the correlation was good under precipitation condition. The root mean square error (RMSE) and correlation coefficient of U-component on 850 hPa varied greatly under precipitation condition, which reflected that the cloud bottom had a certain influence on detection of U-component by wind profiler. The errors of U and V-component wind above 500 hPa were relatively large. (2) Compared with different precipitation conditions, the U-component wind of wind profiler was smaller than sounding data under rainfall and thunderstorm weather condition, and the average error was relatively small and correlation was better under rainfall condition. During shower (snow shower) processes, the error range on different heights were large. The absolute value of average error on 700 hPa was the biggest under thunderstorm condition, followed by 850 hPa. In other precipitation cases, the error of the upper layers was basically greater than that of the lower layers. The V-component wind of wind profiler was larger than sounding data under snowfall condition, and the RMSE below 700 hPa was smaller than that under rainfall, shower (snow shower) and thunderstorm condition, while the RMSE above 700 hPa was the opposite. The error on 700 hPa under thunderstorm condition was large and the correlation was poor. The variation range of the RMSE and correlation coefficient on each height level were small under rainfall condition.

Key words: wind profiler data, horizontal wind, different precipitation conditions, radiosonde, evaluation

摘要： 利用北京市观象台2014—2018年风廓线仪资料和同期探空及MICAPS地面资料，对不同降水条件下风廓线仪水平风与探空资料的误差和相关性进行分析，为风廓线仪资料在降水预报中的深入应用提供参考。结果表明：（1）总体上，各高度层风廓线仪的U分量风均比探空资料偏小；有降水时U分量风误差小且相关性好，其在850 hPa的均方根误差和相关系数均有较大变化，说明云底对风廓线仪探测U分量风有一定影响。500 hPa以上U、V分量风误差较大。（2）风廓线仪探测的U分量风在降雨和雷阵雨时均比探空资料偏小，且在降雨时误差相对较小，相关性较好；阵雨（雪）时不同高度误差变幅大；雷阵雨时700 hPa平均误差绝对值最大，850 hPa次之，其他降水情况基本是高层误差大于低层。降雪时V分量风比探空资料偏大，且700 hPa高度以下均方根误差比降雨、阵雨（雪）和雷阵雨时小，700 hPa以上则相反；雷阵雨时700 hPa误差大，相关性较差；降雨时均方根误差、相关系数在各高度层变化幅度较小。

关键词: 风廓线仪, 水平风, 不同降水情况, 探空, 评价