Sediments from the same river are typically thought to reflect a common provenance, even though they may record notably different, and even opposite, sediment supply and erosion signals. Such discrepancies can introduce significant bias into our interpretation of geomorphological processes. This study focuses on the Yarlung Tsangpo, the Tibetan headwaters of the Brahmaputra River, and contrasts mineralogical and geochemical information on suspended load in transit and on fluvial bars to reveal a major discrepancy in sediment budgets calculated with petrographic and isotopic data for sand and mud fractions. Our results show that in both suspended load and fluvial bars, sand records an overwhelming contribution from the Lhasa block, whereas mud reflects dominant supply from the Himalayan belt. Detritus from the Lhasa block is twice as abundant as Himalayan detritus in sandbars, but Himalayan contribution is 1.4 times that of the Lhasa block in suspended load. Overall, Himalayan sedimentary rocks are estimated to generate as much sediment as Lhasa granitoid and volcanic rocks. Himalayan erosion rates and sediment yields are greatly underestimated if only sand, representing a subordinate part of the total sediment flux, is considered in sediment-budget calculations. Both sand-rich bedload and mud-rich suspended load must be given full consideration in the study of sediment-generation processes. Our results highlight the potential pitfalls of relying solely on sand-sized sediment in provenance analysis and force a reevaluation of how sediment yields and erosion patterns are assessed.
Read full abstract