Abstract

Recently, a tailored gravity field model was developed to fit local terrestrial gravity data by integrating Global Gravitational Models (GGMs), terrestrial gravity data, and Digital Elevation Models (DEMs). The numerical analysis of the newly developed tailored gravity model showed a substantial improvement by means of its possible application for geophysical exploration by exhibiting known geological features over the Southern Benue Trough of Nigeria. In this study, we apply a similar technique to develop a tailored gravity field model at the Limpopo Province in South Africa using a total of 8,603 terrestrial gravity measurements. Validation of results indicates that our tailored gravity model could reproduce the observed gravity data with the accuracy specified by a standard deviation of 8.9 mGal and with a systematic bias less than 0.1 mGal within the study area. We then inspected a possibility of using our tailored gravity field model to improve the accuracy of existing geoid/quasi geoid models at the study area. For this purpose, we compute a new (quasi)geoid model by applying the Remove-compute-restore numerical technique that treats separately the detailed gravity pattern that is closely correlated spatially with the topographic relief, the higher-to-medium gravity signal that is mostly captured by local/regional gravity data, and the long-wavelength gravity signal that is modelled by using GGMs. The accuracy of the new (quasi)geoid model was assessed by using the most recent South African gravimetric quasi-geoid model CDSM09A and the latest hybrid quasi-geoid model of South AfricaSAGEOID10. The comparison of our quasi-geoid model with the CDSM09A and SAGEOID10 quasi-geoid models was done at 7,225 quasi-geoid grid points. The comparison revealed that our new quasi-geoid model closely agrees with the CDSM09A and SAGEOID10 models. The differences between our and CDSM09A quasi-geoid models vary within-0.31 and 0.70 m, with a mean of 0.05 m and a standard deviation of 0.12 m. The corresponding differences between our and SAGEOID10 quasi-geoid models are between-0.35 and 0.70 m with a mean of 0.06 m and a standard deviation 0.12 m. The numerical analysis revealed that the new tailored gravity model could efficiently be used in various geophysical and geodetic applications.