Efficacy of in situ and meteoric 10Be mixing in fluvial sediment collected from small catchments in China

Using measurements of in situ and meteoric Be-10 in fluvial sand to measure erosion rates, quantify soil loss, and trace sediment sources and sinks relies on the assumption that such sediment is well-mixed and representative of the upstream area. We test this assumption at 13 river junctions in three tributary watersheds (200-2500 km(2)) to the Mekong River, Yunnan, China, where human alteration of the landscape is significant and widespread. We find that two of the three watersheds mix well for in situ Be-10 and none mix well for meteoric Be-10 when considering the concentration of Be-10 at the outlet compared to the area-weighted mean of headwater samples. We also assessed mixing at 13 river junctions by comparing the erosion rate-weighted isotopic concentration of sediment taken from tributaries upstream of a junction to the concentration in a sample taken downstream of the junction. With this metric, mixing is generally poor for both in situ and meteoric Be-10 but is better for in situ Be-10 than for meteoric Be-10 (p < 0.05). This is likely because in situ Be-10 is measured in quartz, which is resilient to physical and chemical breakdown in river systems whereas meteoric Be-10 is measured in grain coatings which can abrade and dissolve. Basins eroding faster (> 100 mm/kyr) tend to mix better than slowly eroding basins. We find no evidence that agricultural land use in sampled basins affects sediment mixing downstream. Mixing improves with increased basin area (particularly > 200 km(2)), increased sampling distance downstream from an upstream junction (> 500 m), and increased difference in size between tributaries (one tributary > 3 times larger than the other). The most important factor affecting mixing efficacy for both in situ and meteoric Be-10 is the fraction of the basin area contributing to the downstream sample that does not contribute to the upstream samples. Junctions with > 2% of the basin area unsampled by upstream samples tend not to mix as well. Our data suggest specific sampling location strategies (such as amalgamation) likely to improve the outcome of fluvial network analysis using cosmogenic nuclides.