LANDSCAPE TRANSIENCE IN THE CLEARWATER AND SALMON WATERSHEDS: INSIGHTS FROM BEDROCK RIVER MORPHOLOGY

Landscape transience in the Inland Northwest of the United States may represent an excellent natural experiment through which to explore both (1) the links between geodynamic and geomorphic processes and (2) the role of rock properties in bedrock river incision processes. These opportunities are available because recent incision along the Clearwater and Salmon Watersheds in central Idaho has increased, potentially due to flexural isostactic uplift following interactions between the lithosphere and Yellowstone plume. The increase of incision in the main stem rivers has sent a wave of incision that is still working through many tributaries in the region. The geologic history is such that individual tributaries are underlain by a single lithology but this lithology varies along the axis of the main stem. The wide range of lithologies (e.g., basalt, granite, gneiss, siltite) present and relatively uniform climate aids in highlighting the role of rock properties in transient adjustment. Here, we present estimates of both the transient timescale and incision depths for individual tributaries. We use erosion rates from previous work and measured steepness values to fit an analytical bedrock river incision model. Model accuracy is quantified with the χ2 misfit function, and a range of slope exponent (n) values are evaluated. Our longest predicted durations of incision indicate that an increase in rock-uplift rates may have occurred before 11 to 13 Ma. Our results further show that using an appropriate slope exponent is necessary to achieve model accuracy, and we find that n is often not unity, as often assumed in studies. Although predicted durations and incision depths can vary locally, they offer clearer patterns at broader scales. The relict erosion rate of certain basalt tributaries seems to be significantly lower (about half) than those measured elsewhere in the watershed. An important finding of this work is that the slope exponent on commonly used river erosion models is lithology dependent. We infer that this is a result of different processes that dominate erosion in the various lithologies.