利用中国区域MERRA-2资料计算ZTD和ZWD的精度分析

Accuracy Analysis of ZTD and ZWD Calculated from MERRA-2 Reanalysis Data over China

  • 摘要: 对流层延迟是影响高精度定位与导航的主要误差之一,也是全球导航卫星系统(global navigation satellite system, GNSS)水汽探测的关键参数。美国航空航天局发布了最新一代的大气再分析资料(MERRA-2资料),其可用于计算高时空分辨率的对流层延迟产品,但是目前尚无文献对利用MERRA-2资料计算天顶对流层延迟(zenith tropospheric delay, ZTD)和天顶湿延迟(zenith wet delay, ZWD)的精度进行分析。因此,联合2015年中国陆态网214个GNSS站ZTD产品和分布于中国区域的87个探空站资料,对利用MERRA-2资料在中国区域计算ZTD/ZWD的精度进行评估。结果表明:(1)以陆态网ZTD为参考值,利用MERRA-2资料积分计算ZTD的年均偏差和均方根误差(root mean square error,RMSE)分别为0.32 cm和1.21 cm,且偏差和RMSE均表现出一定的季节变化,总体上呈现为夏季精度低、冬季精度高;在空间分布上,偏差随纬度和高程的变化趋势并不明显,但RMSE随纬度和高程的增加总体上呈现递减的趋势。(2)以探空站计算的ZWD和ZTD数据为参考值,利用MERRA-2资料积分计算的ZWD和ZTD的年均偏差值分别为0.37 cm和0.53 cm,RMSE分别为1.44 cm和1.61 cm,而ZWD/ZTD的偏差和RMSE也表现出一定的时空特性,其时空特性与陆态网GNSS ZTD产品的检验结果相似。由此表明利用MERRA-2再分析资料在中国区域计算ZTD/ZWD具有很高的精度和良好的稳定性,建议可作为中国区域高精度、高分辨率对流层延迟模型构建及高精度GNSS水汽探测的数据源。

     

    Abstract:
      Objectives  Tropospheric delay is one of the main errors in high-precision positioning and navigation, and it is a key parameter in the process of retrieving precipitable water vapor from global navigation satellite system (GNSS) signals. The latest generation of atmospheric reanalysis data of National Aeronautics and Space Administration, MERRA-2, can be used to calculate high-resolution tropospheric delay products. However, there is no literature on the accuracy of the zenith tropospheric delay(ZTD) and zenith wet delay(ZWD) calculated from MERRA-2 data. Therefore, the purpose is to evaluate ZTD and ZWD calculated from MERRA-2 reanalysis data over China.
      Methods  Based on the data of ZTD products from 214 GNSS stations of the crustal movement observation network of China (CMONOC) and 87 radiosonde stations data in 2015, the accuracy and performance of ZTD/ZWD derived from MERRA-2 data in China is evaluated.
      Results  The results show that the ZTD calculated from MERRA-2 data with the annual bias and root mean square error(RMSE)of 0.32 cm and 1.21 cm when compared with CMOMOC-derived ZTD, respectively; the bias and RMSE have obvious seasonality, which generally show relative lower accuracy in summer and higher accuracy in winter.In terms of spatial distribution, the variation trend of bias with latitude and elevation is not obvious, but the RMSE generally shows a decreasing trend with the increase of latitude and elevation. The ZWD and ZTD calculated from MERRA-2 data has the annual bias of 0.37 cm and 0.53 cm compared with radiosonde profiles, respectively; while those of RMSE are 1.44 cm and 1.61 cm, respectively. The bias and RMSE of ZWD and ZTD calculated from MERRA-2 data also have the temporal and spatial variation characteristics, which has similar feature with the tested result of CMONOC GNSS ZTD products.
      Conclusions  Therefore, the ZTD and ZWD calculated from MERRA-2 reanalysis data over China have significant accuracy and stability, which can be used as a data source to establish high-precision and resolution tropospheric correction model and high-precision GNSS water vapor detection in China.

     

/

返回文章
返回