부산대학교 도서관

Cosmogenic nuclides such as 10Be have been widely used to assess present denudation ratesat the catchment scale. A growing number of studies also use this tool to analyzemainly continental sedimentary archives and reconstruct past denudation on largeintegrated area. Because of recent analytical improvements, this tool is now alsoamenable in low concentrated and fine-grained marine sediments, providing theopportunity to study numerous well dated archives throughout the world. However, theuse of 10Be on such fine grained sediments relies on several assumptions: i) thesediments have to be well mixed to be representative of the whole basin, ii) theimpact of the grain size of sediments on the 10Be concentration is either negligible orcan be linked to geomorphological processes, such as landslides, and iii) the 10Besignal is stable from one year to the next, meaning that the catchment is at steadystate. Here, we analyze the evolution of the in situ 10Be cosmogenic concentration in acatchment of the Southern French Alps (drainage area of 2 800 km2, mean elevation of 1 250m, max elevation of 3 200m and mean slope of 23˚) with a short transfer zone and a highvariability of geomorphic, climatic and geological parameters; thus providing a good settingto test those assumptions. Eleven sediment samples have been collected in November 2016 from active riverbeds ofthe Var River and its main tributaries (Tinée, Vésubie, Estéron and Coulomp) and four keylocations have been re-sampled in November 2017. In order to test the impact of grain size on 10Be concentration on fine sediments,samples have been sieved in two fractions: 50-100 μm and 100-250 μm and theirrespective quartz contents isolated. Only two samples points, located in the uppercatchments, yield 10Be concentrations that are not in agreement within uncertainties.On the other hand, grain size has no impact in downstream sediments, notably atthe Var’s outlet, ranging from 4.02±0.78 x 104 atoms/gquartzto 4.40±0.64 x 104atoms/gquartz. We check the mixing of sediments at three majors river junctions by comparing the 10Beconcentration measured downstream of the junction to the sedimentary flux weightedby 10Be concentration measured for each tributary upstream of the junction. Forall three junctions, the expected 10Be concentrations are in agreement with themeasured 10Be concentrations. We consequently assume complete sediment mixing forthe three major river junctions in the Var catchment (Var-Tinée, Var-Vésubie andVar-Estéron). We checked the annual variability of the 10Be signal for four key locations comparingNovember 2016 - when high precipitations events induced high discharge value at theVar’s outlet (< 200 m3/s) - with November 2017 - when low precipitation eventsinduced a much lower discharge value at the Var’s outlet (around 10 m3/s). We findthat despite the high variability of discharge, the denudation rates of the wholewatershed calculated in 2016 and in 2017 are mostly in agreement. Only the upper Varcatchment shows significant differences, with higher erosion rates during the "dry" 2017. [ABSTRACT FROM AUTHOR]