Provenance of Late Permian sediments ofthe Laingsburg subbasin (SW Karoo Basin, South Africa)

FIEDLER, K. and ADELMANN, D.

Sedimentological and petrographic investigations were performed in the upper part of the Ecca Group (Collingham, Vischkuil, Laingsburg, Fort Brown and Waterford Fms.) and the lowermost part of the Beaufort Group (Abrahamskraal Fm.) within the Laingsburg subbasin in the southwestern part of the Karoo Basin (South Africa). They consider the nature and location of the source areas delivering their detritus during Late Permian time.The petrographic analyses of almost 50 fine- to medium-grained sandstone samples cover the lithic arkose and the feldspathic litharenite fields on the QFL diagram (after Folk 1974). There is no significant compositional variation within the entire investigated succession. Quartz, feldspar and fragments of sedimentary, magmatic, and metamorphic rocks with some mica flakes are the main constituents of the sand fraction. Stable heavy minerals include isolated grains of zircon, rutile, garnet, apatite, and tourmaline. The arenites contain less than 10 per cent fine matrix (< 20 µm).
Quartz is the most abundant mineral. Many of the clasts are single-grained, unstrained and mostly free of inclusions. Grain contacts are often concavo-convex, and overgrowths are common. The feldspar clasts consist of plagioclases as well as potassium feldspars. Plagioclase is readily recognized by its polysynthetic twinning. Mineralogical analysis indicates that the average plagioclase composition is either oligoclase or albite. The high Na-content is presumably a result of albitization during diagenesis of the sandstone. Occasionally, the potassium feldspars show a simple Carlsbad twinning. Other feldspar components are microcline with its typical crosshatched pattern, and myrmecite, an intergrowth of feldspar and vermicular quartz. The condition of the feldspars ranges from fresh to altered. Partially their grains are dissolved and replaced by calcite, clay minerals or sericite. First cycle rock fragments, representing a wide range of lithologies, are also common. Mostly they have a sedimentary or metasedimentary origin. Argillite-shale lithic fragments, mainly consisting of illite, sericite and chlorite, predominate with an average amount of more than 50 % of all lithic fragments. Normally, they have irregular outlines conforming to the boundaries of surrounding grains due to compaction of the newly formed sediment. Other minor sedimentary and metasedimentary lithics are silt and sandstone fragments as well as cherts and sericite schists. The most easily recognizable volcanic grains consist of relatively large lath-shaped feldspar crystals set in a finer-grained, partially altered groundmass. The latter is composed of chlorite or illite.
 The composition of detrital sand size components of the investigated succession allows conclusions reflecting the tectonic setting of the provenance areas. Two different provenance types can be distinguished: a recycled orogen source and an additional magmatic source respectively.Strong influence of a recycled orogenic source is documented by a high content of argillaceous lithic fragments, sand and siltstone fragments as well as low-grade metamorphic rock fragments. Additionally, the QpLvLs diagram (after Dickinson 1985) shows the dominance of sedimentary lithic fragments and polyquartz. Therefore, it confirms a source from sedimentary and metasedimentary strata. Paleocurrent data confirm a main sediment transport from southwestern directions for the submarine fan sediments and from southern directions for the deltaic strata. Most likely, the recycled orogenic source was the rising juvenile Cape Fold Belt that comprises the Early Paleozoic sediments of the Cape Supergroup. This fold-and-thrust belt evolved in response to repeated tectonic shortening, starting in the Late Carboniferous and terminating in the Late Triassic (Hälbich et al. 1983).The high content of plagioclase and potassium feldspars (including relatively fresh microcline), and the occurrence of volcanic lithic fragments of a first sedimentary cycle propose a sediment contribution from an additional magmatic source area. Moreover, many rhyolitic and andesitic tuff layers within the Collingham Formation, some basal parts of the Vischkuil Formation, and at the Waterford/Abrahamskraal boundary are evidence of volcanic activity, presumably in the same source area, that provided the mentioned above. Finally, the QmFLt ratios, predominantly plotting in the dissected arc fields indicated by Dickinson (1985), support also the assumption of a magmatic source delivering its detritus into the Laingsburg subbasin. Plate tectonic scenarios by Johnson (1991) and others presume a magmatic arc. Since there is no evidence of such an arc in southwestern Africa, it was probably located to the south of the present Cape Fold Belt, presumably in South America (Ramos & Ramos 1979). Permo-Triassic rhyolitic rocks in close relation to granodiorites, tonalites, and quartz diorites of the plagioclase-rich Paleozoic-Cenozoic batholith belt of Chile could be the source rocks. In this case the Cape Fold Belt did not form a substantial barrier to the distribution of detritus from the igneous and plutonic rock source during Permian time.
 This consideration about the source area in the southwestern part of the Karoo Basin conflicts partially with investigations in the southeastern part. They are various opinions about the origin of its debris. Kingsley (1981) and others believe that the material in the southeast was derived solely from the erosion of a low-grade metamorphic and sedimentary source. He argued employing the abundance of undulatory quartz, the predominance of low-temperature plagioclase, and the rare volcanic fragments. He excluded a volcanic origin of the feldspars, because of their high Na-content. This phenomenon, also observed during this study, is interpreted by a diagenetic albitization of the feldspars. In this case, the material could also be derived from a volcanic or plutonic source. This proposal corresponds to that of many other authors. Johnson (1991), for example, suggested an active magmatic arc as the exclusive source area for the southeastern Karoo Basin. Based on the abundance of volcanic rock fragments and feldspars, he postulated that neither the uplifted Cape Fold Belt nor a gneiss/granite basement constituted the main source for the Ecca and Lower Beaufort strata. An influence of the Cape Fold Belt can be noticed during the Lower Triassic for the first time. The difference between the observations of Johnson (1991) and this study, made in the southwestern part of the Karoo Basin, can be explained as follows. (1) The distance between the magmatic source area and the present southwestern Karoo Basin was significantly larger than in its southeastern part. (2) Only the western part of the Cape Fold Belt was uplifted during the early orogenic activities in the Early Permian (278 ± 5 Ma, Hälbich et al. 1983). Deposition of recycled sediments was restricted to this area. The uplift of the eastern Cape Fold Belt did not start before deposition of the Upper Beaufort Group.

References
Dickinson, W. R. 1985. Interpreting provenance relations from detrital modes of sandstones. In: Provenance of arenites (Edited by Zuffa, G.G.) pp333-361. D.Reidel Publishing Company, Dordrecht, Boston, Lancaster.
Folk, R. L. 1974. Sandstones. Encyclopedia Britannica, Macropedia 16, 212-216.
Hälbich, I. W., Fitch, F. J. & Miller, J. A. 1983. Dating the Cape orogeny. Special Publication Geological Society South Africa 12, 149-164.
Johnson, M. R. 1991. Sandstone petrography, provenance and plate tectonic setting in Gondwana context of the southeastern Cape-Karoo Basin. South African Journal Geology. 91(2/3), 137-154.
Kingsley, C. S. 1981. A composite submarine fan-delta-fluvial model for the Ecca and Lower Beaufort Groups of Permian age in the eastern Cape Province, South Africa. Transactions Geological Society South Africa  84, 27-40.
Ramos, E. E. & Ramos, V. A. 1979. Los Ciclos Magmaticos de la Republica Argentina. Acta VII Congreso Geologia Argentina, 1, 771-786.