Detecting variable rates of erosion using in situ cosmogenic 14C and 10Be

Past changes in climate may have caused the pace of erosion to vary trough time in many landscapes, but constraining such variability remains a key challenge in geomorphology. A range of surface processes operating in fluvial landscapes can potentially be affected by climate, for example via changes to vegetation patterns, average precipitation rates, and storm frequency. To quantify the variability, requires the ability to answer at least the following first-order question: Was erosion during glacial periods faster or slower than in the current interglacial period? In this theoretical study, we show how combined measurements of in situ 10Be and 14C can be used to answer this question in some fluvial landscapes. However, the mean rate of erosion and the amplitude of the change are critical factors. For example, a change in erosion rate by a factor of 3 or larger on glacial-interglacial timescales should be detectable by combined analysis of 10Be and 14C in fluvial landscapes, providing that they generally erode at rates of 10-5-10-4 m/yr . In contrast, similar changes in erosion rate cannot be detected if the landscape erodes more slowly (<10-5m/yr), as the corresponding error in the erosion rate from 14C is then too large. Changes in erosion rate are also difficult to detect if the landscape erodes at higher rates (>10-4m/yr), as the timescales covered by the two nuclides are then too similar. We present a general sensitivity study that constrains the required conditions for detecting variable rates of erosion on glacial/interglacial time scales. Our study highlights the potentials of combined measurement of in situ 10Be and 14C in fluvial landscapes, and it offers theoretical predictions that may be useful when devising sampling strategies to study the effect of climate on erosion in fluvial landscapes.