In the doctoral dissertation we deal with the modeling of the vertical gravity gradient. The knowledge of the vertical gradient of the gravity field is of great importance in gravimetry and consequently in geodesy, since the gravity acceleration measured with the instrument must be reduced to a stabilized point. Based on the given geodetic and geophysical data, we can create various combinations of models from which we calculate the vertical gravity gradients. We are interested in the accuracy of the modeled values. For this purpose, we measured the vertical gravity gradients at 44 test points using Scintrex CG-3M and CG-5 relative gravimeters, processed the measurements and obtained reference values. We use these to empirically test the effects of individual elements in the modeling process on the final results. Based on the deviations between the measured and modeled values, we were able to perform analyzes and statistical evaluations of the results. This allows us to determine the optimal model and data selection criteria for modeling vertical gravity gradients. We also compared values calculated from some global models of the Earth’s gravity field created using satellite technology. For the purpose of modeling, a new model of the density of topographic masses of Slovenia was created based on existing geological data. It was used for modeling of vertical gravity gradients. After several decades, new maps of gravimetric anomalies of the Republic of Slovenia were created based on new and improved input data, including a new map of Bouguer anomalies. For this purpose, an analysis of the existing gravimetric data was performed and some gross errors were removed from the data sets.