TY - JOUR AU - Jaime GARBANZO-LEÓN AU - Alonso VEGA FERNÁNDEZ AU - Mauricio VARELA SÁNCHEZ AU - Juan Picado SALVATIERRA AU - Robert W. KINGDON AU - Oscar H. LÜCKE PY - 2020/07/29 Y2 - 2024/03/28 TI - A regional Stokes-Helmert geoid determination for Costa Rica (GCR-RSH-2020): computation and evaluation JF - Contributions to Geophysics and Geodesy JA - Contrib. Geophys. Geod. VL - 50 IS - 2 SE - original research papers DO - 10.31577/congeo.2020.50.2.3 UR - https://journal.geo.sav.sk/cgg/article/view/237 AB - GNSS observations are a common solution for outdoor positioning around the world for coarse and precise applications. However, GNSS produces geodetic heights, which are not physically meaningful, limiting their functionality in many engineering applications. In Costa Rica, there is no regional model of the geoid, so geodetic heights (h) cannot be converted to physically meaningful orthometric heights (H). This paper describes the computation of a geoid model using the Stokes-Helmert approach developed by the University of New Brunswick. We combined available land, marine and satellite gravity data to accurately represent Earth's high frequency gravity field over Costa Rica. We chose the GOCO05s satellite-only global geopotential model as a reference field for our computation. With this combination of input data, we computed the 2020 Regional Stokes-Helmert Costa Rican Geoid (GCR-RSH-2020). To validate this model, we compared it with 4 global combined geopotential models (GCGM): EGM2008, Eigen6C-4, GECO and SGG-UM-1 finding an average difference of 5 cm. GECO and SGG-UM-1 are more similar to the GCR-RSH-2020 based on the statistics of the difference between models and the shape of the histogram of differences. The computed geoid also showed a shift of 7 cm when compared to the old Costa Rican height system but presented a slightly better fit with that system than the other models when looking at the residuals. In conclusion, GCR-RSH-2020 presents a consistent behaviour with the global models and the Costa Rican height systems. Also, the lowest variance suggests a more accurate determination when the bias is removed. ER -