The effects of basin asymmetry on sand composition: Examples from Lake Tanganyika, Africa

Distinctive compositional trends occur in sand suites collected from structurally diverse segments of Lake Tanganyika (East African Rift System). Climate (subhumid, tropical) and source lithology (mixed granitic and metasedimentary rocks) are generally invariant between the source areas studied. Therefore, differences in sand composition relate directly to differences in compositional modification during transport and deposition of the sands. Sand composition along contrasting segments of the lake basin can be related to the strong structural asymmetry of the basin. We point-counted medium sands from river mouth, deltaic, beach, and shallowwater environments of four structurally distinct, lake-margin settings: (1) escarpment margin (Pemba, Zaire site); (2) hinged margin (Nyanza Lac, Burundi, and Rumonge, Burundi sites); (3) accommodation-zone margin (Magara, Burundi site); and (4) axial margin (Ruzizi River site, Burundi). These four margin-types are typical of the structural morphology in strongly asymmetric half-graben rift basins. Using the Gazzi-Dickinson point counting methodology, margins two through four cluster tightly in QFL space, suggesting similar source lithology. Samples from the Pemba site (escarpment margin) are significantly depleted in feldspar. Because of the uncertainties in interpreting this difference, the escarpment margin sands were not included in subsequent analyses. Variations in sand composition occur between the remaining sites when using the traditional point-counting methodology. The suite of fluvial sands entering the lake basin along the accommodation-zone margin (Magara site) averages 38% rock fragments compared to an average of 15 to 20% and 18% rock fragments for the suite of fluvial sands entering the basin along the hinged or axial margins, respectively. This is a fundamental contrast that correlates directly to differences in drainage basin size. Reworking at the depositional site produces additional compositional differences across the structural margins. In comparing the sample suites for each site, it is the variability in the data that is more important than average composition in highlighting these differences. Minimal compositional modification occurs at the depositional site along both the accommodation-zone margin and the axial margin. In contrast, significant environmental overprinting occurs along the hinged margin sites. Limited compositional modification along the accommodation-zone margin results from the steep depositional gradient and the very narrow high-energy zone which limits reworking of the sediment. The limited reworking and the consequent minimal compositional modification along the axial margin are primarily a function of the high sedimentation rate. The compositional modification (enrichment in quartz) along the hinged margin is a function of low sediment accumulation rates coupled with low subsidence rates, leaving large regions of shallow-water substrate exposed to long periods of reworking. Chemical alteration of labile phases also occurs along the hinged margin through carbonate replacement of feldspar. Our results suggest that, with careful statistical analysis, it is possible to use sandstone composition to distinguish between tectonic margins typical of half-graben basins. Using this approach in ancient studies, it is necessary to collect from several coeval facies among different study sites and to constrain source lithology and climate. In addition, these results are most appropriate for humid-climate basins, where weathering processes are most extreme.