广东风廓线雷达水平风产品质量分析

Quality Analysis of Horizontal Wind Products from Wind Profilers in Guangdong

关键词:  风廓线雷达产品  再分析数据  同址探空数据  误差垂直分布

Key words:  wind profiler product  reanalysis data  the same-site sounding data  vertical distribution of errors

摘要:

为检验风廓线雷达质控产品的质量，本文利用ERA5再分析数据和同址探空数据，分别从误差和相关性统计、晴/雨天的表现、时间演变、误差垂直分布等几方面对2023年广东省风廓线雷达的小时水平风产品开展了对比评估。结果表明，与验证了合理性的ERA5再分析数据相比，风廓线雷达的风向误差较大，但风速、U风和V风和ERA5结果接近，并且无论是晴天还是雨天，其平均误差都在±0.5m·s-1内、平均绝对误差和均方根误差在1.7~3.3m·s-1之间、相关性超出0.7，雨天误差略大于晴天，但雨天相关性也更大，表明降雨时期的风廓线雷达数据也具有较强的合理性；和同址探空数据对比，汕头风廓线雷达的风速、U风、V风在2000m以上的平均绝对误差和均方根误差垂直分布比较平滑稳定，基本分布在1~5m·s-1之间，以U风误差最大；多部风廓线雷达的垂直误差分布还表明在垂直探测模式切变高度容易出现误差峰值与突变。总体而言，广东风廓线雷达质控后的水平风产品具有较高的可用性。

Abstract:

To examine the quality of horizontal wind products from wind profilers, this study uses ERA5 reanalysis data and sounding data from the same site to evaluate the hourly horizontal wind products from wind profilers in Guangdong Province in 2023, by using error and correlation statistics, evaluating the performance on sunny/rainy days, and evaluating time evolution and vertical distribution of errors. Compared with ERA5 reanalysis data that have been verified by the sounding measurements, the errors of wind direction in wind profiler products are large, whereas the errors of wind speed, U and V are small. Whether it is sunny or rainy, the average errors of wind speed, U and V in wind profiler products are within ± 0.5 m · s-1, the average absolute errors and root mean square errors are 1.7-3.3 m · s-1, and the correlation coefficients exceed 0.7. The errors on rainy days are slightly larger than those on sunny days, but the correlation between wind profiler products and ERA5 reanalysis on rainy days is also greater, indicating that the wind profiler products during rainy periods are also reasonable. Compared with sounding data at the same site, Shantou, the vertical distribution of the average absolute errors and root mean square errors of wind speed, U and V above 2 000 m is relatively smooth and stable, basically distributed between 1-5 m · s-1, with U having the largest errors. The vertical error distributions of several wind profiler radars also indicate that error peaks and abrupt changes are prone to occur at the shear heights of vertical detection mode. In general, the horizontal wind products of wind profilers have relatively high quality.