Comprehensive analysis of interpolation methods for modeling the spatial distribution of gravity anomalies

Modeling the spatial distribution of gravity anomalies requires accounting for numerous factors that influence the accuracy of results. The primary factors include the interpolation methods used to construct regular grids of gravity anomalies, as well as the distribution and number of observation points. This study compares the accuracy of various interpolation methods for gravity anomalies based on the WGM2012 model. The analysis utilized gravity anomaly data obtained from 200 GNSS stations located in Ukraine and 355 test points of a hypothetical gravimetric network. The research aimed to evaluate the accuracy of interpolation methods such as Inverse Distance to a Power, Kriging, Minimum Curvature, Moving Average, Nearest Neighbor, Polynomial Regression, and Radial Basis Function in tasks of modeling the spatial distribution of anomalies using data from the WGM2012 model. The analysis was performed based on calculated differences between interpolated and original values, supported by graphical and statistical data. The results allowed for the classification of interpolation methods by accuracy: 1) High accuracy with uniform value distribution; 2) Moderate accuracy with a balanced distribution; 3) Low accuracy with large amplitude variations. The study demonstrated that a well-founded selection of an interpolation method can significantly enhance the accuracy of modeling the spatial distribution of gravity anomalies and provide reliable results for solving geophysical problems.