Investigation of the contribution of terrestrial vertical gravity gradient data to regional gravimetric geoid modelling

In this study, using free-air gravity anomaly derived from terrestrial gravity measurements, vertical gravity gradient (VGG) of the free-air gravity anomaly is modelled by Heiskanen/Moritz and Least Squares Collocation (LSC) methods. Modelled vertical gravity gradient of the free-air anomaly is added to normal vertical gravity gradients, thus modelled vertical gravity gradients are obtained by means of aforementioned two different methods. Modelled VGG are compared with terrestrial VGG measurements measured at 159 locations in the Western part of Turkey. The Heiskanen/Moritz method and LSC methods show 339 Eötvös and 188 Eötvös standard deviation coherence respectively with measured VGG. The gravimetric quasi-geoid calculated with Molodensky approach and LSC method show 2,9 cm standard deviation agreement with GPS/levelling derived quasi-geoid. Moreover, the impact of the vertical gravity gradient of the free-air gravity anomaly in the calculation of the separation term for the conversion of gravimetric quasigeoid-to-geoid, is investigated. Our results show the importance of taking into consideration of this term, especially in mountainous areas, showing differences reaching up to 5 cm. In the Poincaré-Prey reduction used for the orthometic height computation, the impact of the vertical gravity gradient of the free-air gravity anomaly to the GPS/levelling geoid heights is studied. This effect is found to be in the same order of magnitude and direction with its impact of the separation term to the geoid. Additionaly, it has been observed that the difference between gravimetric geoid and GPS/levelling derived geoid effects neutralize each other. For this reason, to reveal the contribution of the vertical gravity gradient of the free-air anomaly to gravimetric geoid determination, GPS/levelling data is found not be appropriate as validation data. In conclusion, either in gravimetric geoid/quasi-geoid determination or in GPS/levelling derived geoid/quasi-geoid determination, for a better accuracy, it is proposed to take the vertical gravity gradient of the free-air gravity anomaly derived from measured or modelled VGG into consideration. In addition, to validate gravimetric geoid, especially in mountaneous areas, alternative methods, besides GPS/levelling techniques needs to be investigated.