Linkage of Normal Heights Obtained from GNSS and Refined XGM2019 GGM to Brazilian Vertical Data Using Different Approaches

Abstract

This work aimed to analyze the use of different approaches to link normal heights obtained via Global Navigation Satellite System (GNSS)/Global Geopotential Model (GGM) refined by the RTM technique to the Brazilian Vertical Data (Imbituba Brazilian Vertical Datum – IBVD and Santana Brazilian Vertical Datum – SBVD). Specifically, it analyzed approaches based on the weighted mean of discrepancies between height anomalies, the zero- level geopotential value, the Geodetic Boundary Value Problem (GBVP) solution, and the use of parametric modeling of a plane with a scale factor. For the numerical tests, two different study regions have been used, the first with heights referenced to IBVD and the second to SBVD. Using the first three approaches, the local modeling idea has been investigated in both regions. In this context, spatial cluster analysis of the outliers of differences between local and global height anomalies defined the sub-regions. In the fourth approach, the treatment of local modeling was initially considered. In the accuracy analysis of linkages, it has been verified that approaches based on the mean of the discrepancies between height anomalies and using zero- level geopotential value propose practically the same results. On the other hand, there were improvements at the centimeter level with the use of the GBPV solution-based approach compared to the first two, except for two worsening cases. With the approach based on parametric modeling, the accuracy results were mainly worse considering the approaches with local modeling. The most significant differences reached the decimeter level.