2019 CSDMS meeting-014
Assessing basin-scale lithologic controls on river and landscape evolution
Bedrock lithology has been shown to strongly influence how rivers and landscapes respond to tectonic perturbations, yet the specific variables and mechanisms that set how lithology controls river erosion are poorly understood. Recent field and modeling work suggests that one important lithologic control on channel response may be the delivery of large, generally immobile boulders from hillslopes to channels. This raises the possibility that differences in boulder delivery rates between lithologies may cause substantial differences in how landscapes respond to tectonics. An intriguing recent study suggested that in the Mendocino Triple Junction (MTJ) region of northern California, bedrock lithology might control the frequency and size of boulders delivered to channels, and therefore govern channel steepness and river evolution (Bennett et al., 2016). We further test this hypothesis here.
The Central Belt of the Franciscan Complex, a mix of sheared graywacke and mudstone, contains large blocks of more resistant serpentinite, greenstone, and amphibolite that are delivered to channels by earthflows. The adjacent Coastal Belt generally lacks such boulders, and sediment delivery to channels is dominated by shallow landsliding. This geologic setting provides a unique opportunity to test whether boulder abundance exerts a first-order control on landscape form. We use a landscape-scale analysis of channel steepness and active width indices, local topographic relief, lithology, and mapped boulder occurrence to understand the differences between the catchments eroding the Central Belt and those eroding the Coastal Belt. We find that channels are steeper in the Central Belt than in the Coastal Belt, both across the whole MTJ region and when averaged over 10-50 km2 subcatchments. Channels are also generally narrower in the Central Belt. This result could reflect lithologic controls or spatial heterogeneity in erosion rates. To control for the latter, we construct clusters of neighboring subcatchments that are free of knickpoints to explore possible controls of lithologic makeup (percent of a subcatchment underlain by Central Belt rocks) on channel steepness independent of erosion rate variations. We find inconsistent relationships between lithologic makeup and channel steepness within a given cluster of catchments with similar baselevel history. Finally, we compared channel segments adjacent to hillslope failures with segments far from failures. Central Belt channels show greater absolute increases in steepness adjacent to hillslope failures, but relative increases in steepness are consistent between the Central Belt and Coastal Belt.
Our preliminary results suggest that Central Belt channels are steeper and narrower than Coastal Belt channels, but that the lithological influence on steepness is difficult to disentangle from the effects of spatially variable erosion rates. We are continuing to map in-channel boulder size distributions to assess the relative importance of intra- vs. inter-lithologic variability in setting boulder concentrations and landscape form.