Item

Abstract

Several geoid models have been developed to convert geodetic heights measured by the Global Navigation Satellite Systems (GNSS) to heights in the Hong Kong Principal Datum (HKPD). Although gravity measurements were used to compile these geoid models, heights of levelling benchmarks in HKPD were determined from precise spirit levelling measurements but without involving gravity data. To address these inconsistencies, the orthometric heights of HKPD levelling benchmarks were determined from levelling and gravity data. Moreover, the new geoid model HKGEOID-2022 was computed and fitted with the geometric geoid heights at GNSS-levelling benchmarks derived from newly determined orthometric heights. Numerical procedures used to prepare the HKGEOID-2022 geoid are discussed in this study. A gravimetric geoid was computed adopting the KTH method. A systematic bias between the gravimetric and geometric geoid heights at GNSS-levelling benchmarks was modelled and reduced by applying a 7-parameter similarity transformation. The accuracy analysis revealed that the resulting detailed geoid model HKGEOID-2022 fits the geometric geoid heights with a standard deviation of ±2.2 cm. This accuracy is compatible with the estimated uncertainties of GNSS measurements as well as with the expected accuracy of a newly developed geoid model, both at the level of approximately ±1–2 cm.