Search

Whole-rock geochemical data and compositional balance analysis (CoBA), combined with a magnetic susceptibility model obtained from a high-precision magnetic survey, are useful exploration tools in the Jiama porphyry–skarn deposit, southern Tibet, China. In this paper, three innovative aspects of the model are described that have assisted exploration of the Jiama deposit: an improved geological understanding of a thrust–nappe structure, hydrothermal fluid pathways, and magma mixing; data- and knowledge-driven compositional balance analysis of whole-rock geochemistry; and a high-precision magnetic survey of the deposit. The local strata provided the materials and natural trap for skarn formation, and a site for fluid circulation and porphyry ore precipitation. An interlayer detachment zone, gliding nappe structure, and cylindrical fissure system provided channels for fluid migration, and the Niumatang Anticline provided the accommodation space for the porphyry system. Miocene magma mixing and crustal contamination led to the development of complex mineralization in the Jiama area. The geochemical signatures obtained from CoBA, and the spatial distribution of the strata, structures, igneous rocks, alteration, mineralization, and degree of erosion enable integration of the geochemical and geological data. The "demagnetization skylight" identified by magnetic anomalies and a magnetic susceptibility model obtained with UBCmag3D software revealed the location of a concealed porphyry system. A magnetic susceptibility of 0.014–0.016 was found to identify the location of an interlayer skarn. Our exploration indicators of the Jiama porphyry–skarn deposit can be used to guide ore exploration throughout the region. • Meaningful geochemical associations are gained from data- and knowledge- driven CoBA. • Using geochemical associations to infer geology, alteration, mineralization and erosional degree of Jiama deposit. • Geophysics contributes to positioning porphyry complex and interlayer skarn. • A comprehensive exploration model involves multi-source information is proposed. [ABSTRACT FROM AUTHOR]