Due to wide spread usage of the satellite positioning techniques especially GPS, we need to precisely determine geoid model in order to use GPS measurements for height determination, as an alternative of traditional leveling techniques in geodetic applications. Precise local geoid modelling using GPS/Leveling data, apart from the existing models such as geopotential models and gravimetric geoid models could be an interesting investigation topic. An important question is, ‘What accuracy level can be achieved using this approach?’ However precession of this modelling could be influenced by some issues such as data quality or modelling techniques. In this paper, we attempt to assess the implementation of modern learning-based computing techniques including artificial neural networks and adaptive network-based fuzzy inference systems compared with multivariate polynomial regression equations in GPS/Leveling Geoid modeling. This assessment carried out in a small and dense network of GPS/Leveling benchmarks in contrast with previous studies, located in shahin-shahr, Isfahan. And these high quality data make it possible to achieve an accuracy of better than 1 cm. The results show a few millimeter superiority of ANN and ANFIS derived geoid models in terms of root mean square error, as well as in terms of coefficient of determination. And RMSE=8cm, R2=0.9949 and RMSE=7cm, R2=0.9964 achieved for this models respectively. Therefore ANFIS derived geoid model provide the most accurate geoid heights in the study area.

Frequency decomposition of the Earth gravitational potential in terms of spherical/ellipsoidal harmonics has been of the significant matter for a wide range of applications such as the geodetic, oceanographic, and geophysical purposes. These days, thanks to the notable advancement in the field of satellite altimetry, monitoring the sea level on the global scale has been realized in practice. Meanwhile, the gravity information may be derived from these valuable data if the accurate mean dynamic topography has been obtainable. In this respect, the mean dynamic topography can be determined via the oceanic or geodetic approaches. In the oceanic manners, the mean dynamic topography is derived using the oceanic information such as salinity, temperature, and surficial currents; while according to the geodetic methods one can obtain the mean dynamic topography by integration of satellite altimetry measurements and global geopotential models. In this contribution, we aim at assessing the feasibility of improving the pre-existing geopotential models by means of satellite altimetry observations. To this end, the sea surface topography is estimated using the geopotential and mean sea level models in a constrained least squares sense. As such, we can arrive at the Gauss-Listing geoid over the sea areas derived from the computed sea surface topography and the known mean sea level values. The Bruns formula is then implemented to reduce the resultant geoid into the gravitational potential values over the oceans on the surface of the reference ellipsoid. On the other hand, the gravitational potential values over the continental regions are obtained on the surface of the reference ellipsoid via the geopotential model of interest, once the topographic bias corrections have been considered. Lastly, a new point-wise geopotential model in terms of spherical harmonics is developed through application of the spherical harmonic analysis to the worldwide gravitational potential. As the case study, the presented methodology has been evaluated so as to improve the two global geopotential models, namely EGM2008 and go_cons_gcf_2_dir, up to the degree and order 90. Accordingly, the DTU10 mean sea level model, which has been derived from information of Topex/Poseidon, ERS1, ERS2, ENVISAT, Geosat, GFO, and Jason satellites, has been applied to the EGM2008 and go_cons_gcf_2_dir models in order to estimate the sea surface topography based on the proposed optimization solution. Consequently, the improved versions of the global geopotential models have been developed thanks to the application of the harmonic analysis to the gravitational potential values that have been attained over the sea and land areas on the surface of the reference ellipsoid. Based upon the numerical results of the assessment of the developed models at the first-order GPS/leveling points within the test areas in Iran and Finland, the capabilities of the proposed method in deriving enhanced geopotential models have been asserted. Moreover, the comparison of the consequential enhanced models with respect to the BGI gravity points have demonstrated the efficiently of the method throughout the world. As a whole, we have deduced that the presented method can be applicable to significantly improve an extensive range of the global geopotential models.