Transformation of sediment grain size distributions by abrasion and fragmentation in an alpine catchment: linking hillslope weathering and erosion to bedload transport and channel incision

River incision into bedrock and thus topographic relief depend on, and also influence, the size distribution of sediment produced on hillslopes and supplied to channels. Sediment size is therefore central to the feedbacks between weathering, sediment production, and channel incision that drive landscape evolution. However, quantifying these feedbacks is challenging due to difficulty in measuring how hillslope sediment size varies at catchment scales and how size distributions evolve as particles are transported through the channel network. Recent work has shown that detrital apatite (U-Th)/He ages measured in all size classes present at a catchment outlet can reveal elevation gradients in sediment size produced on hillslopes. We compared this approach with in-situ grain size measurements on hillslopes at Inyo Creek, which spans 2 km of relief in the Sierra Nevada, California, USA. We find that the two independent data sets do not agree. For example age data suggest that boulders originate at lower elevations whereas sand is produced uniformly across the catchment. The hillslope data show the reverse: spatially uniform boulder production with sand only from lower elevations. This divergence may reflect particle wear during transport from hillslope sources to the catchment outlet; only boulders produced near the outlet arrive at the sampling site intact, whereas sampled sand is a mix of sand produced by particle wear and by weathering on hillslopes.

To explore this hypothesis, we developed a numerical model to predict particle wear and the evolution of grain size distributions by fragmentation and attrition, mechanisms that occur in steep catchments. The model is calibrated with data from two laboratory experiments: individual rock drops to quantify the probability of fragmentation on impact and number of fragments produced as functions of impact energy; and bulk sediment tumbling in 4 rotating drums ranging from 0.2 to 4.0 m in diameter. The wear model can reproduce the variation in wear rates and evolution of grain size distributions in the rotating drum experiments. Wear rates also correlate with energy expenditure in drums of different sizes, permitting extrapolation to the field. We incorporate the wear model into a population balance model to track the evolution of particle size distributions due to wear, mixing of sediments from sources across the catchment, and the transfer of (U-Th)/He ages among sediment sizes. A plausible range of modeled wear rates yields a match between predicted and measured age distributions by size class at the catchment outlet. This holistic picture of the life cycle of sediment grain size links the upstream geomorphic conditions that control hillslope weathering and erosion with the downstream size and abundance of coarse grained tools that control channel incision into bedrock.