{"title":"Sub-lithospheric mantle sources for overlapping southern African Large Igneous Provinces","authors":"L. Ashwal","doi":"10.25131/SAJG.124.0023","DOIUrl":null,"url":null,"abstract":"\n At least four spatially overlapping Large Igneous Provinces, each of which generated ∼1 x 106 km3 or more of basaltic magmas over short time intervals (<5 m.y.), were emplaced onto and into the Kaapvaal Craton between 2.7 and 0.18 Ga: Ventersdorp (2 720 Ma, ∼0.7 x 106 km3), Bushveld (2 056 Ma, ∼1.5 x 106 km3), Umkondo (1 105 Ma, ∼2 x 106 km3) and Karoo (182 Ma, ∼3 x 106 km3). Each of these has been suggested to have been derived from melting of sub-continental lithospheric mantle (SCLM) sources, but this is precluded because: (1) each widespread heating event sufficient to generate 1 to 2 x 106 km3 of basalt from the Kaapvaal SCLM (volume = 122 to 152 x 106 km3) would increase residual Mg# by 0.5 to 2 units, depending on degree of melting, and source and melt composition, causing significant depletion in already-depleted mantle, (2) repeated refertilization of the Kaapvaal SCLM would necessarily increase its bulk density, compromising its long-term buoyancy and stability, and (3) raising SCLM temperatures to the peridotite solidus would also have repeatedly destroyed lithospheric diamonds by heating and oxidation, which clearly did not happen. It is far more likely, therefore, that the Kaapvaal LIPs were generated from sub-lithospheric sources, and that their diverse geochemical and isotopic signatures represent variable assimilation of continental crustal components. Combined Sr and Nd isotopic data (n = 641) for the vast volumetric majority of Karoo low-Ti tholeiitic magmatic products can be successfully modelled as an AFC mixing array between a plume-derived parental basalt, with <10% of a granitic component derived from 1.1 Ga Namaqua-Natal crust. Archaean crustal materials are far too evolved (εNd ∼ -35) to represent viable contaminants. However, a very minor volume of geographically-restricted (and over-analysed) Karoo magmas, including picrites, nephelinites, meimechites and other unusual rocks may represent low-degree melting products of small, ancient, enriched domains in the Kaapvaal SCLM, generated locally during the ascent of large-volume, plume-derived melts. The SCLM-derived rocks comprise the well-known high-Ti (>2 to 3 wt.% TiO2) magma group, have εNd, 182 values between +10.5 and -20.9, and are characteristically enriched in Sr (up to 1 500 ppm), suggesting a possible connection to kimberlite, lamproite and carbonatite magmatism. These arguments may apply to continental LIPs in general, although at present, there are insufficient combined Sr + Nd isotopic data with which to robustly assess the genesis of other southern African LIPs, including Ventersdorp (n = 0), Bushveld (n = 55) and Umkondo (n = 18).","PeriodicalId":49494,"journal":{"name":"South African Journal of Geology","volume":" ","pages":""},"PeriodicalIF":1.2000,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"South African Journal of Geology","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.25131/SAJG.124.0023","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"GEOLOGY","Score":null,"Total":0}
引用次数: 5
Abstract
At least four spatially overlapping Large Igneous Provinces, each of which generated ∼1 x 106 km3 or more of basaltic magmas over short time intervals (<5 m.y.), were emplaced onto and into the Kaapvaal Craton between 2.7 and 0.18 Ga: Ventersdorp (2 720 Ma, ∼0.7 x 106 km3), Bushveld (2 056 Ma, ∼1.5 x 106 km3), Umkondo (1 105 Ma, ∼2 x 106 km3) and Karoo (182 Ma, ∼3 x 106 km3). Each of these has been suggested to have been derived from melting of sub-continental lithospheric mantle (SCLM) sources, but this is precluded because: (1) each widespread heating event sufficient to generate 1 to 2 x 106 km3 of basalt from the Kaapvaal SCLM (volume = 122 to 152 x 106 km3) would increase residual Mg# by 0.5 to 2 units, depending on degree of melting, and source and melt composition, causing significant depletion in already-depleted mantle, (2) repeated refertilization of the Kaapvaal SCLM would necessarily increase its bulk density, compromising its long-term buoyancy and stability, and (3) raising SCLM temperatures to the peridotite solidus would also have repeatedly destroyed lithospheric diamonds by heating and oxidation, which clearly did not happen. It is far more likely, therefore, that the Kaapvaal LIPs were generated from sub-lithospheric sources, and that their diverse geochemical and isotopic signatures represent variable assimilation of continental crustal components. Combined Sr and Nd isotopic data (n = 641) for the vast volumetric majority of Karoo low-Ti tholeiitic magmatic products can be successfully modelled as an AFC mixing array between a plume-derived parental basalt, with <10% of a granitic component derived from 1.1 Ga Namaqua-Natal crust. Archaean crustal materials are far too evolved (εNd ∼ -35) to represent viable contaminants. However, a very minor volume of geographically-restricted (and over-analysed) Karoo magmas, including picrites, nephelinites, meimechites and other unusual rocks may represent low-degree melting products of small, ancient, enriched domains in the Kaapvaal SCLM, generated locally during the ascent of large-volume, plume-derived melts. The SCLM-derived rocks comprise the well-known high-Ti (>2 to 3 wt.% TiO2) magma group, have εNd, 182 values between +10.5 and -20.9, and are characteristically enriched in Sr (up to 1 500 ppm), suggesting a possible connection to kimberlite, lamproite and carbonatite magmatism. These arguments may apply to continental LIPs in general, although at present, there are insufficient combined Sr + Nd isotopic data with which to robustly assess the genesis of other southern African LIPs, including Ventersdorp (n = 0), Bushveld (n = 55) and Umkondo (n = 18).
期刊介绍:
The South African Journal of Geology publishes scientific papers, notes, stratigraphic descriptions and discussions in the broadly defined fields of geoscience that are related directly or indirectly to the geology of Africa. Contributions relevant to former supercontinental entities such as Gondwana and Rodinia are also welcome as are topical studies on any geoscience-related discipline. Review papers are welcome as long as they represent original, new syntheses. Special issues are also encouraged but terms for these must be negotiated with the Editors.