DO FAULT SCARPS ACTUALLY "DIFFUSE?"

Fault scarps are meter-scale locations of relief created by surface-rupturing earthquakes. Fault scarps hold a unique place in the history of hillslope geomorphology because they have straightforward initial conditions and are modified by the same processes as hillslope. In fact, the original models of model for hillslope evolution focused on diffusivity applications to fault scarps. However, diffusion-based conceptions of hillslope sediment transport process may not be the best representation of fault scarp evolution. The steep nature of fault scarps may instead lend themselves more towards formulations incorporating the concept of nonlocality, where far-field locations upslope of a point contributes to the sediment flux at said point. In this presentation, we review predictions of both diffusion based and nonlocal models of fault scarp topographic evolution. We compare these data with a well-preserved field site at Deep Creek, Utah and with a data compilation of published fault scarp topographic profiles. Through this comparison, we find that fault scarp topographic evolution shows a signature of nonlocal sediment transport that is especially pronounced for fault scarps that record a single rupture and is subtly present for much older scarps with multi-rupture histories. When compared with field site data, the nonlocal model produces a notably more realistic simulation of fault scarp evolution; both in depositional patters within the accommodation space formed from earthquakes and in the scarp profile form overall. As scarp-like features are very common in hillslopes, particularly at the interface between hillslopes and channels, the finding that nonlocality best explains scarps may have major implications for our theories of hillslope topographic evolution.