Quantifying quartz enrichment and its consequences for cosmogenic measurements of erosion rates from alluvial sediment and regolith

In-situ cosmogenic 26Al and 10Be record the residence time of quartz grains near the earthʹs surface, and thus can be used to measure whole-catchment erosion rates averaged over millennial time scales. Quartz is enriched in hillslope regolith by the dissolution of more soluble minerals; thus, its residence time will be longer than the regolith average. It has been noted that this introduces a bias into erosion rate estimates derived from cosmogenic nuclide concentrations in regolith or alluvium [Geomorphology 27 (1999) 131], but the magnitude of this bias has not previously been measured. The enrichment of quartz in regolith, and the resulting bias in cosmogenic erosion rate estimates, can be quantified using concentrations of immobile elements (such as zirconium) in bedrock and regolith. Here we show that the erosion rate bias introduced by regolith dissolution is less than 12%, across 22 granitic catchments that span a wide range of temperate climates. Except in extreme weathering environments, biases due to regolith dissolution will be a small component of the overall uncertainty in cosmogenic erosion rate measurements.