One significant geochemical data processing aim is to delineate anomalies associated with mineral deposits. In areas with strong surface weathering, the accumulation centres of surface geochemical anomalies are often not completely matched with locations of mineral deposits. This affects anomaly interpretation and mineral prospectivity prediction. In order to solve this challenging problem, quantitative prediction of mineral prospectivity based on multi-information fusion techniques has been one of the research hotspots in the field of data analysis in recent years. This study first summarized the geological background and metallogenic control factors of each tectonic unit in Guangxi, and then analysed the relationship between Pb–Zn deposits and Pb–Zn geochemical anomalies from 60 767 geochemical stream sediment samples. Based on the re-classified geochemical element contents, gravity, aeromagnetic data and fault, magmatic rock, magmatic rock and fault intersection buffer data as input layers, together with 302 Pb–Zn ore occurrences selected as training data sets, quantitative prediction of prospectivity for Pb–Zn ore deposits in the study area was carried out using back-propagation neural network and fuzzy weights-of-evidence methods. It was found that the Pb–Zn mineral prospectivity prediction areas based on multi-information fusion techniques can eliminate effectively the influence of secondary accumulation of elements during weathering of carbonate rocks on the recognition of deposit-associated stream sediment geochemical anomalies, and identify effectively the mineral resources closely related to rock mass and structure distribution. These analyses reveal the metallogenic regularity of Pb–Zn deposits from the perspective of data mining based on machine learning and geographical information system multi-information fusion for delineation of prospective metallogenic target areas. The purpose here was to provide new ideas for reducing the effects of secondary weathering of extensive carbonate rocks in Guangxi, and in other regions with similar landscapes, on mineral prospectivity prediction.

Thematic collection: This article is part of the Applications of innovations in geochemical data analysis collection available at:

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.