Comparison of Ray-Tracing Packages for Troposphere Delays
Comparison of Ray-Tracing Packages for Troposphere Delays
复制标题
DOI：
10.1109/tgrs.2011.2160952
复制
发表时间：
2012-02-01
影响因子：
8.2
通讯作者：
Gegout, Pascal
中科院分区：
文献类型：
--
作者：
Nafisi, Vahab;Urquhart, Landon;Gegout, Pascal
2010年上半年，在国际大地测量学协会4.3.3工作组的主持下，开展了一项比较活动，以评价和比较不同射线追踪软件的对流层延迟，有五个机构参加：GFZ德国地球科学研究中心、空间大地测量研究小组、国家信息和通信技术研究所，新玩法大学和维也纳理工大学大地测量和地球物理研究所。向比较活动的参与者提供了欧洲中期天气预报中心对筑波站（日本）和韦策尔站（德国）的业务分析的高分辨率数据。这些数据包括相对于平均海平面，温度和比湿度的位势差，所有在等压面。此外，还提供了关于大地水准面起伏的信息，与会者计算了5年的射线追踪总延迟。仰角和方位角的每一度。在一般情况下，我们发现良好的协议之间的射线跟踪倾斜因子从不同的解决方案在5。如果根据ECMWF的相同压力水平数据确定，则高度。标准差和偏差在1 cm水平（或对于某些组合明显更好）。其中一些差异是由于算法和插值方法的差异。如果与从ECMWF本地模型水平数据确定的倾斜因子相比，偏差可能会显着更大。
A comparison campaign to evaluate and compare troposphere delays from different ray-tracing software was carried out under the umbrella of the International Association of Geodesy Working Group 4.3.3 in the first half of 2010 with five institutions participating: the GFZ German Research Centre for Geosciences (GFZ), the Groupe de Recherche de Geodesie Spatiale, the National Institute of Information and Communications Technology (NICT), the University of New Brunswick, and the Institute of Geodesy and Geophysics of the Vienna University of Technology. High-resolution data from the operational analysis of the European Centre for Medium-Range Weather Forecasts (ECMWF) for stations Tsukuba (Japan) and Wettzell (Germany) were provided to the participants of the comparison campaign. The data consisted of geopotential differences with respect to mean sea level, temperature, and specific humidity, all at isobaric levels. Additionally, information about the geoid undulations was provided, and the participants computed the ray-traced total delays for 5. elevation angle and every degree in azimuth. In general, we find good agreement between the ray-traced slant factors from the different solutions at 5. elevation if determined from the same pressure level data of the ECMWF. Standard deviations and biases are at the 1-cm level (or significantly better for some combinations). Some of these discrepancies are due to differences in the algorithms and the interpolation approaches. If compared with slant factors determined from ECMWF native model level data, the biases can be significantly larger.