Comprehensive and strict validation of neutral atmosphere ray tracing algorithms and some improvements for space geodetic radio signals

Abstract

This study presents a more thorough and rigorous method for the validation of atmospheric ray tracing algorithms used for computing the path delays of space geodetic microwave signals. Romberg integration through a hypothetical atmospheric structure, e.g. the US standard atmosphere, is recommended to derive the reference values of slant total delay (STD) and geometric delay (GD) for validations. Enlighten by the validation results, simplification for the ray tracing algorithm APM-3D and revision for the piecewise linear (PWL) method are proposed and both are proved to be valid. Comparison results show that in a spherically symmetric atmosphere (one-dimensional (1D) media), if the atmosphere is vertically stratified with sufficient height layers, all 1D, two-dimensional (2D), and three-dimensional (3D) ray tracing algorithms tested yield identical STDs and GDs with reference values at an elevation angle of 5°; while with limited height layers, the revised PWL method exhibits better performance than other 1D algorithms considered. The 2D and 3D algorithms are further compared using realistic 3D atmosphere from ERA5 reanalysis data, and it indicates mm to cm level of differences in GDs; while the STD differences range between <0.1 mm to cm level, depending on weather condition and whether shooting is implemented during the tracing. Both STD and GD discrepancies become larger under severe weather, and the magnitude of 2D-3D difference is found to be consistent with the azimuthal variation of STDs. In comparison of results obtained without shooting, shooting procedure tends to decrease the STD differences but has little impact on the differences in GD. The simplified APM-3D can replace the original version with 10−5 mm level of agreements in STD.