A note of caution on the use of boulders for exposure dating of depositional surfaces

Abstract

Exposure dating of boulders has been widely applied to determine the age of depositional surfaces under the assumption that the pre-depositional nuclide component in most boulders is negligible. Here we present a case study on fluvial terraces at the active mountain front of the eastern Andes, where this assumption is clearly invalid, because sandstone boulders (n=13) from terraces at two sites contain a highly variable inherited 10Be component and have apparent 10Be ages that exceed the age of the respective surface by up to ~90ka. Likewise, boulders from active stream channels (n=5) contain a substantial inherited 10Be component, equivalent to 5-48ka of exposure. The age of the fluvial terraces is well determined by two approaches that allow to correct for the pre-depositional nuclide component: 10Be dating of amalgamated pebbles and 10Be depth profiles on sand samples. At site 1, three terraces have 10Be ages of 3-5ka (T2), 11-13ka (T3), and 16-20ka (T4), which are consistent with the terrace stratigraphy. The age of terrace T3 is confirmed by a calibrated 14C age of 12.61±0.20ka BP obtained from a wood sample. At site 2, terrace T3 has a 10Be age of 13-16ka. The average inherited 10Be concentration of sand grains - determined from depth profiles and stream sediments - is small and equivalent to 1-3ka of exposure. In contrast, the mean inheritance of pebbles and boulders is higher and equivalent to exposure times of ~10ka and ~30ka, respectively. These differences in the pre-depositional nuclide component are related to the different provenance and transport history of sand, pebbles, and boulders. The sand is derived from rapidly eroding Miocene sediments exposed near the mountain front, whereas the pebbles and boulders originate from Triassic sandstones in the internal part of the fold-and-thrust belt. On their way to the mountain front, boulders and pebbles were temporarily stored and irradiated in alluvial fans that are currently reworked. As sediment deposition in intramontane basins and their subsequent excavation is common in the Andes and other fold-and-thrust belts, the presence of pre-depositional nuclide components should be evaluated when applying exposure dating at active mountain fronts.