Denudation rates and a topography-driven rainfall threshold in northern Chile: Multiple cosmogenic nuclide data and sediment yield budgets

The quantification of geomorphic process rates on the outcrop- and the orogen-scale is important to describe accurately the interaction between the relative effects of erosion, tectonics and climate on landscape evolution. We report single and paired cosmogenic nuclide (10Be, 26Al and 21Ne) derived erosion rates and exposure ages on hillslope interfluves from the tectonically active western central Andes that show a distinct spatial variation. A positive correlation of erosion rates with elevation and present-day rainfall rates is observed. Erosion rates at lower altitudes–the hyperarid Coastal Cordillera and the Western Escarpment with the northern part of the Atacama Desert–are extremely low and of the order of 10–100 cm/My (nominal exposure ages 1–6 My). In contrast, erosion rates at higher altitudes–the semiarid Western Cordillera–range up to 4600 cm/My (nominal exposure ages 0.02–0.1 My). This latter average long-term bedrock erosion rate record, suggested to be coupled to an orographically controlled pattern of rainfall, is also reflected in the pattern of denudation rates derived from a short-term decadal record of limited sediment yield data. Specifically, denudation rates calculated from sediment flux data are of a similar order of magnitude as erosion rates deduced from long-lived cosmogenic nuclides from bedrock hillslope interfluves of the Western Cordillera. Nevertheless, the production and the supply of sediment from the western Andean slope are very limited.Analysis of multiple cosmogenic nuclides allows simultaneous determination of erosion rates and exposure ages but also reveals complex exposure histories of non-bedrock samples, such as boulders or amalgamated clast samples. Notably, this study shows that saturation of nuclides, usually assumed in studies where only a single nuclide is analyzed, is rather the exception than the rule, as revealed by erosion island plots. Constant erosion that started much later than the formation age of the rocks or episodic erosion by spalling can partially explain non-steady-state concentrations and more complicated exposure scenarios. Furthermore, the use of multiple nuclides with different half-lives allowed us to infer that no significant variations in long-term erosion rates have occurred and that at the Western Escarpment erosion rates have been low and constant for most of the late Neogene. Nevertheless, the time intervals necessary to reach steady-state concentrations for cosmogenic nuclides can be quite different from those needed for landscapes to reach steady state.