Search

The DTRF2014 solution is the latest ITRS realization of DGFI-TUM computed in the framework of ITRF2014. It is mainly characterized by two innovations: it is the first secular ITRS realization considering non-tidal loading corrections derived from geophysical models. Secondly, the DTRF2014 release includes all information necessary to approximate the observed instantaneous station positions at any epoch of the observation time span. Therefore, besides the classical SINEX and EOP files, the applied non-tidal loading corrections, the residual time series of station positions as well as the translation time series of the DTRF2014 origin are provided. The DTRF2014 is computed from the same input as ITRF2014 comprising the full history of observation data of the four space geodetic techniques Very Long Baseline Interferometry (VLBI), Satellite Laser Ranging (SLR), Global Navigation Satellite System (GNSS), and Doppler Orbitography and Radiopositioning Integrated by Satellite (DORIS). The DTRF2014 solution is computed using the DGFI-TUM approach based on the combination of normal equation systems. With reference to the IERS Conventions, the origin of DTRF2014 is realized from SLR input data only. The scale of the DTRF2014 solution is assumed to be statistically equal for SLR and VLBI. This assumption is based on tests performed in the framework of the DTRF2014 computation showing a range for the scale offset at epoch 2000.0 of ± 3.3 mm. For the GNSS and the DORIS subnetwork of the DTRF2014 solution, a weighted mean scale of SLR and VLBI is realized. Cross-validations performed between DTRF2014 and ITRF2014 show the high consistency of present ITRS realizations. For GNSS, VLBI and SLR the obtained transformation parameters range within ± 1.7 mm for positions (except for the VLBI and SLR scale and the VLBI z-translation) and ± 0.3 mm/yr for the velocities, the parameters for DORIS are all below 4.2 mm and 0.25 mm/yr (except for the scale rate). The RMS of the transformations, reflecting the agreement of network geometry, range between 0.55 mm for VLBI and 2.25 mm for DORIS. The RMS of station velocities is between 0.08 and 0.62 mm/yr. The DTRF2014 solution provides a high precision and reliability confirmed by various validations performed internally and by external groups. For instance, applications of DTRF2014 for a precise orbit determination (POD) of the altimetry satellite TOPEX and Jason-2 show small radial crossfit residuals between SLR and DORIS and indicate a good performance of DTRF2014. [ABSTRACT FROM AUTHOR]

In a previous paper, Glomsda et al. (2020) revisited the impact of distinct parts of non-tidal loading in the analysis of geodetic Very Long Baseline Interferometry (VLBI). The loading is represented by displacements of the reference positions of the observing VLBI antennas, which are variables of a corresponding Gauss-Markov model for estimating various geodetic target parameters. These displacements were applied at two different levels of the model, the observation and the normal equation level, and quite similar results were obtained for both cases. In this paper, the authors provide a more detailed theoretical discussion of the application of site displacements at the distinct levels, which also comprises the a posteriori application at the solution level. For each case, the respective formulas and implications for the Gauss-Markov model are derived, and equations for assessing the differences between the estimated parameters are established. In this way, the authors aim to create a deeper understanding of the results of the previous paper, which show the capability of the normal equation level to approximate the application of site displacements at the observation level with less effort and prerequisites, and to provide evidence for the claims made in that paper (for VLBI only): (i) the chosen reference frame of the site displacements is basically irrelevant except for the solution level; (ii) the Jacobi matrix does not change substantially; (iii) the loss of temporal resolution of the site displacements is more important than the linear approximation of the functional model at the normal equation level; and (iv) the station coordinate estimates for all three levels strongly depend on the regularizing (datum-) conditions of the model. The theoretical results are substantiated with numerical examples, which consider site displacements generated from non-tidal atmospheric loading by the Earth-System-Modelling group of the Deutsches GeoForschungsZentrum (GFZ). However, the results are valid for site displacements of any source. [ABSTRACT FROM AUTHOR]