Development of a Geostatistical Co-Kriging Framework for Integrating Geophysical and Geochemical Data in Mineral Prospectivity Mapping

This study develops an original geostatistical co-kriging framework for quantitatively integrating geophysical and geochemical survey data in mineral prospectivity mapping across the Nigerian basement complex, contributing a new approach to exploration data integration that exploits spatial correlation structures. Mineral prospectivity mapping combines multiple evidence layers from different exploration methods to rank the target potential of exploration areas. However, conventional index overlay and logistic regression methods treat data layers independently, ignoring the spatial cross-correlation between geophysical and geochemical anomalies that carries additional information about co-located mineralisation. Geostatistical co-kriging provides a statistically rigorous framework for integrating spatially correlated datasets that exploits cross-variogram structures. This study applies the framework to a multi-dataset exploration database for a 5,000 square kilometre area in Ondo and Ekiti States comprising aeromagnetic data, ground IP-resistivity surveys, NGSA stream sediment geochemistry for gold, arsenic, and lead, geological mapping, and known mineral occurrence locations. Cross-variogram analysis characterises the spatial correlation structure between pairs of exploration datasets. Co-kriging prediction and simulation produce probabilistic mineral prospectivity maps at 250-metre resolution. Validation uses prospectivity prediction against known mineral occurrences withheld from the training dataset. Findings reveal significant cross-variogram correlation between aeromagnetic intensity and arsenic geochemistry (sill correlation 0.62) and between IP chargeability and gold geochemistry (sill correlation 0.74), confirming spatially coupled geophysical-geochemical signals. The co-kriging prospectivity map improves known occurrence prediction recall by 34 percent over single-method mapping and by 22 percent over uncorrelated overlay methods. The study recommends the co-kriging framework as a standard data integration methodology for Nigerian basement complex mineral exploration.

Keywords: geostatistical co-kriging, mineral prospectivity mapping, geophysical integration, Nigerian basement complex, exploration data.