Lead isotopic studies of some Late Proterozoic stratabound ores of central Africa

Abstract The preliminary results of a lead isotopic study of the Owambo basin-type ore association are presented. This association occurs in the Late Proterozoic Damaran, Katangan and West Congolian fold belts of central Africa, and is confined to platform sequences that border the Angola—Congo craton. Two mineralogically distinct types of mineralisation constitute the ore association. The Berg Aukas type is mineralogically similar to the lead-rich members of the Mississippi Valley-type ore association. Kabwe (formerly Broken Hill, Zambia) and Berg Aukas (Namibia) are the most important occurrences of this type included in this study. Tsumeb and Kombat (Namibia), Kipushi (Zaire) and M'Passa (Congo Republic) are the most important studied representatives of the mineralogically more complex, copper-rich Tsumeb-type mineralisation. Mineralisation of both types occurs in stratabound and discordant palaeokarst and solution-collapse structures within carbonate sequences, adjacent to basement highs. The Tsumeb-type mineralisation also occurs as stratabound and discordant bodies in sandstones. Both types of mineralisation occur in close proximity to each other and to a third type, the Copperbelt-type stratiform copper mineralisation. However, all three types occur in separate mineralised, regionally extensive stratigraphic horizons. The lead- and sulphur-isotope and geochemical evidence place constraints on a number of variables. These include the ore fluid migration paths, metal sources, the relationship between different mineralisation types and the timing of the mineralisation. The evidence suggests that the ore fluids were related to basin dewatering and that they accumulated metals from their aquifer rocks during this process. Sulphides were precipitated in karst and other permeable horizons within evaporite-bearing sequences. The Tsumeb-type mineralisation is associated with unconformities which coincide with a major change from shallow marine carbonate to molasse sedimentation. This mineralising event was probably broadly contemporaneous throughout the three Pan-African fold belts. Data from the Owambo basin suggest that this occurred about 600 Ma ago. All of the galena lead isotopic trends are considered to have been derived by mixing processes, and therefore have no direct age significance. The following conclusions are suggested by the data: 1. (1) Tsumeb- and Berg Aukas-type mineralisation of the Owambo basin may have a common origin. Rock lead data suggest that they have a mixing component of shale-derived lead in common. Phosphatic oolites and arkosic sediments might have provided other components but these rocks have not yet been investigated. 2. (2) The Tsumeb-type trend of the Owambo basin is distinctly more radiogenic northwards from the shale basin. Fluid migration over considerable distances is implied; the mines studied are up to 70 km apart. 3. (3) Tsumeb-type and unclassified occurrences of the Kakontwe basin, Zaire (Kipushi, Lombe and Kengere) display similar isotopic relationships and probably represent a single mineralising event. The Mazabuka prospect, Zambia, may also belong to this group. This suggests that fluids migrated out of the basin on a 200 km (and possibly more than 600 km) dewatering front. 4. (4) The northern and southern groups of Tsumeb-type mineralisation in the West Congolian display different lead isotopic trends but occur in comparable stratigraphic positions. This suggests dewatering from (or via) separate basins towards a single basement high (the Mbanzu Ngungu ridge). 5. (5) The Bukanda (Calcaire rose: K.s. 1.2.1), Ndola (Katontwe Limestone) and Kiseba (basal Roan) Pb—Zn prospects were not thought to be related before this study was commenced. However, their similar isotopic relationships on both the uranogenic and uranothorogenic plots suggests a common origin. This would require a mineralising event which permeated the entire Katangan sequence. 6. (6) Data from the Owambo basin support such stratigraphically penetrative mineralising events. Berg Aukas-type mineralisation which shows a single isotopic trend occurs throughout a more than 1 km stratigraphic thickness of the Otavi Group. Isotopically distinct Tsumeb-type mineralisation penetrated a similar stratigraphic thickness within a single mine (Tsumeb). Solution pipes and karsted fault zones were probably important in conveying solutions through impermeable horizons within the stratigraphy. 7. (7) The Berg Aukas-type leads of Berg Aukas and Kabwe are isotopically identical although they occur in different fold belts. One possible explanation for this would be a major lead mixing component in common which was derived from an arkosic aquifer of Kibaran provenance. 8. (8) Stratiform copper mineralisation is a feature of the sandstones which overlie the Tsumeb-type mineralisation in each of the fold belts. Early pyrite has been replaced by copper minerals in one such occurrence. This feature has been reported from a number of stratiform copper provinces throughout the world. The pyrite was found to have a lead isotopic composition similar to its host arenite (in which it replaces iron-bearing minerals). Data from a sample of the galena that is sparsely distributed through this mineralisation was found to plot on the lead isotopic trend of the underlying, isotopically distinct, Tsumeb-type mineralisation. The data obtained at this stage of the study are consistent with a model which involves discrete dewatering events from entire depositional systems (i.e., on the scale of subbasins). This is supported by recent data from the Mississippi Valley district, where values corresponding to 30% of the length of the regionally extensive J-type lead isotopic trend have been found in single galena crystals (Hart et al., 1981). The interdependence of sulphur and lead isotopic ratios in such ore requires the mixing of at least two separate lead sources. Separate sulphur sources have also been proposed in this model (i.e. sulphur and lead transported in the same solution). This is consistent with the data but not essential to the model. The mixing of two lead-bearing fluids to varying degrees prior to reaching a sulphur source might produce fluids of varying lead isotopic composition and oxygen fugacity that could give the observed result. The central African geochemical, sulphur isotopic and geological evidence is consistent with a single sulphur source near the site of deposition. Two other observations emerge from this study. The first is that care should be taken when interpreting mineralisation in thinner Phanerozoic sequences. Mineral occurrences of slightly different type in host rocks of significantly different age and composition are commonly interpreted as being related to separate mineralising events (e.g., the Mississippi Valley district deposits and the lead sandstones and carbonate-hosted zinc of the southeastern Massif Central; Bjorlykke and Sangster, 1981). However, single mineralising events in the central African province affect thick stratigraphic sequences. These could span more than one geological Period if present in younger rocks. Evidence for such a young and stratigraphically penetrating mineralising event is suggested by the Mississippi Valley lead isotopic data (i.e., the Cambrian of southeast Missouri versus the Pennsylvanian of the Tri-State district). This is also suggested by recent lead isotope data from the Massif Central (e.g., Largentiere versus Les Malines; Brevart et al., 1982). The second observation is that Copperbelt-type stratiform copper mineralisation may have a similar origin to that proposed for all of the other types of mineralisation discussed above. It might sometimes have formed from the same fluids. This is suggested by the lead isotopic data and is supported by observations elsewhere (e.g., in the Michigan copper from palaeoaquifers in the Portage Lake volcanics and Copper Harbour conglomerate into stratiform mineralisation at the base of the overlying Nonesuch shale (Butler and Burbank, 1929; Brown, 1971). This is consistent with a dewatering rather than a syngenetic origin.