Gravity anomalies play critical roles in geological analysis, geodynamic monitoring, and precise geoid modeling. Obtaining accurate gravity data is challenging, particularly in inaccessible or sparsely covered regions. This study evaluates machine learning (ML) methods—Support Vector Regression (SVR), Gaussian Process Regression (GPR), and Ensemble of Trees—for predicting gravity anomalies in southeastern Kazakhstan and compares their effectiveness with traditional Kriging interpolation. A dataset, consisting of the simple Bouguer anomaly values, latitude, longitude, elevation, normal gravity, and terrain corrections derived from historical maps at a scale of 1:200,000, was utilized. Models were trained and validated using cross-validation techniques, with performance assessed by statistical metrics (RMSE, MAE, R²) and spatial error analysis. Results indicated that the Exponential GPR model demonstrated the highest predictive accuracy, outperforming other ML methods, with 72.9% of predictions having errors below 1 mGal. Kriging showed comparable accuracy and superior robustness against extreme errors. Most prediction errors from all methods were spatially associated with mountainous regions featuring significant elevation changes. While this study demonstrated the effectiveness of machine learning methods for gravity anomaly prediction, their accuracy decreases in complex terrain, indicating the need for further research to improve model performance in such environments.

This study presents the digitization and integration of Kazakhstan’s legacy gravimetric maps at a scale of 1:200,000 into a modern geospatial database using ArcGIS. The primary objective was to convert analog gravity data into a structured, queryable, and spatially analyzable digital format to support contemporary geoscientific applications, including geoid modeling and regional geophysical analysis. The project addresses critical gaps in national gravity coverage, particularly in underrepresented regions such as the Caspian Sea basin and the northeastern frontier, thereby enhancing the accessibility and utility of gravity data for multidisciplinary research. The methodology involved a systematic workflow: assessment and selection of gravimetric maps, raster image enhancement, georeferencing, and digitization of observation points and anomaly values. Elevation data and terrain corrections were incorporated where available, and metadata fields were populated with information on the methods and accuracy of elevation determination. Gravity anomalies were recalculated, including Bouguer anomalies (with densities of 2.67 g/cm³ and 2.30 g/cm³), normal gravity, and free-air anomalies. A unified ArcGIS geodatabase was developed, containing spatial and attribute data for all digitized surveys. The final deliverables include a 1:1,000,000-scale gravimetric map of free-air gravity anomalies for the entire territory of Kazakhstan, a comprehensive technical report, and supporting cartographic products. The project adhered to national and international geophysical mapping standards and utilized validated interpolation and error estimation techniques to ensure data quality. The validation process by the modern gravimetric surveys also confirmed the validity and reliability of the digitized historical data. This digitization effort significantly modernizes Kazakhstan’s gravimetric infrastructure, providing a robust foundation for geoid modeling, tectonic studies, and resource exploration.