小波去噪的广西加权平均温度插值研究

针对广西地区探空站稀少,难以获得精确的T_m问题,GGOS atmosphere提供了利用ECMWF的相关资料计算而得到的时间分辨率为6h(UTC 00:00:00,06:00:00,12:00:00,18:00:00)、空间分辨率为2.5°×2°的全球T_m格网数据可以在没有气象数据的情况下获得较高时空分辨率的T_m,该文利用GGOS atmosphere T_m格网数据对广西地区4个探空站插值T_m,并用无线电探空数据计算的T_m检验其精度;对误差进行分析后,选取最优小波基与尺度对其残差去噪,利用去噪后得到的曲线建立T_m的改正模型。实验结果表明,插值T_m经基于小波去噪的模型改正后,其RMSE为1.29K;Bevis模型的RMSE为10.71K;GPT2_1W模型的RMSE为3.56K;改正模型精度优于传统模型,可以达到地基GPS反演GNSS-PWV的精度要求。

Study of weighted mean temperature interpolation of the atmosphere grid data based on wavelet denoising in Guangxi

Aiming at the problem that it is difficult to obtain accurate Tm becase of rare radiosonde satations distributed in Guangxi,GGOS atmosphere can provide a time resolution of 6 hours(at UTC 00:00:00,06:00:00,12:00:00,18:00:00) and a spatial resolution of 2.5° × 2°gridded data including the weighted mean temperature,which can be obtained without weather data by using GGOS atmosphere grid.So GGOS atmosphere grid data were used after using bilinear interpolation method and wavelet de-noising to calculate the accuracy statistical results of four stations in Guangxi at the same station and same time,comparing with the true value,the residuals was used after wavelet de-noising to build the correction model of Tm,and it was used to forecast the 2015's Tm in Guang xi.Comparing with Bevis model and GPT2 _ 1W model,the result of the correction model have high precision and reliability,which can meet the reqirement of GNSS-PWV inversion.