Cosmogenic isotopes, short‐lived radionuclides, elemental concentrations and thermochronometric indicators are measured in river sand to quantify erosion rates and trace sediment sources, and/or infer erosional processes. Interpretations of detrital sediment analyses are often based on the rarely tested assumption of time‐invariant tracer concentration. A better understanding of when and where this assumption breaks down and what sampling strategies minimize temporal and small‐scale spatial variance will improve science done using detrital river sediment. Here, we present new and previously published spatial and temporal replicates measured for in situ and meteoric 10Be (10Bei and 10Bem, respectively). Our new data include 113 replicate pairs, taken from agricultural and/or tectonically active watersheds in China months to millennia apart and spatial replicates taken up to 2 km apart on the same day. The mean percentage difference is 10% (−122% to 150%) for both systems considered together; the mode is close to 0% for both systems; and 36% of pairs of samples replicate within our analytical accuracy at 2σ. We find that 10Bei replicates better than 10Bem (p < 0.01). 10Bei replicability is worse in steeper basins, suggesting that stochastic processes (i.e. landslides) affect reproducibility. 10Bem replicability is worse in larger basins, suggesting non‐conservative behavior of 10Bem as sediment moves downstream. Our results are consistent with the few previously published replicate studies. Considering all replicate data in a wide range of landscapes, in areas with deep erosional processes, replicability is poor; in other areas, replicability is good. This suggests that, in steep, tectonically active, and/or agricultural landscapes, individual detrital sediment measurements do not represent upstream rates as well as larger populations of samples. To ensure that measurements are representative of the upstream watershed, our data suggest that samples be amalgamated either over time or from several places close by in the same channel. Copyright © 2017 John Wiley & Sons, Ltd.

Agricultural land use doubled sediment loads in western China's rivers Author information removed for double-blind review.

Erosion rates of tropical landscapes are poorly known. Using measurements of in situproduced 10 Be in quartz extracted from river and landslide sediment samples, we calculate longterm erosion rates for many physiographic regions of Panama. We collected river sediment samples from a wide variety of watersheds (n = 35), and then quantified 24 landscape-scale variables (physiographic, climatic, seismic, geologic, and land-use proxies) for each watershed before determining the relationship between these variables and long-term erosion rates using linear regression, multiple regression, and analysis of variance (ANOVA). We also used grainsize-specific 10 Be analysis to infer the effect of landslides on the concentration of 10 Be in fluvial sediment and thus on erosion rates. The strongest and most significant relationship in the dataset was between erosion rate and silicate weathering rate, the mass of material leaving the basin in solution. None of the topographic variables showed a significant relationship with erosion rate at the 95% significance level; we observed weak but significant correlation between erosion rates and several climatic variables related to precipitation and temperature. On average, erosion rates in Panama are higher than other cosmogenically-derived erosion rates in tropical climates including those from Puerto Rico, Madagascar, Australia and Sri Lanka, likely the result of Panama's active tectonic setting and thus high rates of seismicity and uplift. Contemporary sediment yield and cosmogenically-derived erosion rates for three of the rivers we studied are similar, suggesting that human activities are not increasing sediment yield above long-term erosion rate averages in Panama.3 10 Be concentration is inversely proportional to grain size in landslide and fluvial samples from Panama; finer grain sizes from landslide material have lower 10 Be concentration than finegrained fluvial sediment. Large grains from both landslide and stream sediments have similarly low 10 Be concentrations. These data suggest that fluvial gravel is delivered to the channel by landslides whereas sand is preferentially delivered by soil creep and bank collapse. Furthermore, the difference in 10 Be concentration in sand-sized material delivers by soil creep and that delivered by landsliding suggests that the frequency and intensity of landslides influences basin scale erosion rates.

Using measurements of in situ and meteoric 10 Be 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 10 Be and only one mixes well for meteoric 10 Be when considering the concentration of 10 Be at the outlet compared to the areaweighted 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 10 Be but is better for in situ 10 Be than for meteoric 10 Be (p < 0.05). This is likely because in situ 10 Be is measured in quartz,

LongNew data from 14 watersheds in the states of Santa Catarina (n = 7) and Rio de Janeiro (n = 7) show that erosion rates vary there from 13 to 90 m/My (mean = 32 m/My; median = 23 m/My) and that the difference between erosion rates of basins we sampled in the two states is not significant. Sampled basin area ranges between 3 and 14,987 km 2 , mean basin elevation between 235 and 1606 m, and mean basin slope between 11 and 29°. Basins sampled in Rio de Janeiro, including three that drain the Serra do Mar escarpment, have an average basin slope of 19°, whereas the average slope for the Santa Catarina basins is 14°. Mean basin slope (R 2 = 0.73) and annual precipitation (R 2 = 0.57) are most strongly correlated with erosion in the basins we studied. At three sites where we sampled river sand and cobbles, the 10 Be concentration in river sand was greater than in the cobbles, suggesting that these grain sizes are sourced from different parts of the landscape.Compiling all cosmogenic 10 Be-derived erosion rates previously published for southern and southeastern Brazil watersheds to date (n = 76) with our 14 sampled basins, we find that regional erosion rates (though low) are higher than those of watersheds also located on other passive margins including Namibia and the southeastern North America. Brazilian basins erode at a pace similar to escarpments in southeastern North America. Erosion rates in southern andsoutheastern Brazil are directly and positively related to mean basin slope (R 2 = 0.33) and weakly but significantly to mean annual precipitation (R 2 = 0.05). These relationships are weaker when considering all southern and southeastern Brazil samples than they are in our smaller, localized data set. We find that smaller, steeper headwater catchments (many on escarpments) erode faster than the larger, higher-order but lower slope catchments. Erosion in southern and southeastern Brazil appears to be controlled largely by mean basin slope with lesser influence by climate and lithology.