Controls and distribution patterns of gold mineralisation in the Manica- Mutare- Odzi greennstone belt (Western Mozambique and Eastern Zimbabwe)

Abstract

The Manica-Mutare-Odzi (MMO) greenstone belt is a late Archaean linear structure that was formed around 2,74 Ga. The belt has produced up to1995 c. 84 tonnes of Au mainly from shear zone/fault hosted and quartz vein hosted deposits. Gold deposits are arranged in corridors, which seem to control their localization. Theses corridors are oriented at an angle to know main orientation of the structures that host mineralization in the belt. The distribution of known gold deposits reflect a pattern of the structures controlling mineralization to an extent that depends on the degree of mining in the MMO belt, and are important requisite to the design of exploration models and better understanding of gold genesis inte Archaen. This study sought to understand the distribution patterns of gold deposits and the controls that influenced the Au minerisalisation localization in the MMO greenstone belt. The study has established criteria for gold deposit recognition, built a database for structuring, easy retrieval, update and manipulation of captured data, it also performed spatial association analysis to generate maps of evidence and finally it produced gold prospectivity maps for the MMO greenstone belt. The study has reached among others the following results: (i)Surface and cancealed structures, granodioritic intrusions and mafic ultramafic rocks of Bulawayan age mainly control mineralization in the study area. (ii) Gold deposits in the study area are fractal. The fractal dimension suggests a single event process or synchronous harmonic set of geological processes acting as one, probably related to hydrothermal systems that transported and deposited gold in favourable sites, governed the gold mineralization distribution. (iii) Autocorrelation analysis shows presence of important gold deposits at the intersections between known and the newly interpreted crypt structures. (iv) According to Fry analyses gold deposits cluster at distance ranges between 2 and 30 km. (v) The shear zones and zones of BIF were successfully mapped using iron parameter from PIMA (Portable Infrared Mass Analyser) data. (vi) The gathered information on controls and distribution patterns of the gold occurrences were spatially related to the area and, weights of evidence were calculated and optimum distance of influence established for each parameter. The cutoff distances are 500 for felsic and intermediate intrusions, 400 for shear zones and quartz veins, 700 for Shamvaian/Bulawayan contact, 900 for BIF, 800 for greenstone/granite contact, 500 for the Fry interpreted lineaments and 900 m for Bulawayan lithologies. (vii) In the contingency table for calculated chi-square two binary predictor patterns (the Fry lineaments and the BIF) did not meet the criteria for conditional independence and they were discarded from the model. (viii) the gold prospectivity model as designed depicts 54% of the deposits within high and very high prosprctivity areas and only 16% of the deposits fall within low prospectivity area. (ix) Using the validation dataset, 51% of the deposits are distributed within the high and very high prospectivity areas and only 20% of the gold occurrences fall within low prospectivity area. (x) The 30% of gold occurrences falling within area of moderate prospectivity index for both designing and validation datasets shows that the model is consistent. (xi) The prospectivity maps highlight areas of high, very high prospectivity which can be used to guide future gold exploration programs