Slant path delay model


It is currently widely accepted that errors in modeling the variable path delay in the neutral atmosphere is the major contributor to the error budget of space geodesy techniques that operate in the radio frequency range: very long baseline interferometry (VLBI), global positioning system (GPS), and Doppler Orbitography and Radiopositioning Integrated by Satellite (DORIS). Therefore, any improvements in modeling the path delay in the neutral atmosphere will improve accuracy of geodetic results.

Recent advances in remote sensing and in computer hardware made it feasible to produce the numerical weather models of the atmosphere with high temporary and spatial resolution. We expect that the accuracy and resolution of these models will improve in the future. The output of these models represents the 4D field of the atmospheric parameters. Therefore, the path delay can be computed directly from this field by numerical integration without using inaccurate regression models.

The motivation of this project is to derive a rigorous and computationally efficient algorithm for computing the slant path delay through the 4D, inhomogeneous, elliptical atmosphere designed for processing space geodesy observations on a global, dense network; to validate this model using space geodesy observation; and to develop a procedure for automatic update of the slant path delays in order to use it for routine data analysis of radioastronomy and space geodesy observations. The algorithm is intended to be used in a service of the slant path delay computation for processing VLBI, GPS, and DORIS observations.


Software package SPD was developed for computing slant path delays by solving differential equations of wave propagation that produce the wave trajectory and integration of path delay along that trajectory.

Slant path delays

Results of slant path delay computation in ascii format for 218 VLBI sites for the period of time from January 01, 1980 through March 31, 2010 using the output of MERRA numerical weather model.

The same slant path delays in binary format are available as well.


... is in progress

Web page was prepared by Leonid Petrov ()
Last update: 2010.02.12_22:36:10