MAGNITUDE-FREQUENCY DISTRIBUTION OF LANDSLIDES IN A PHYSICAL EXPERIMENT

In incising fluvial landscapes, sidewall slope angles increase, stability decreases, and landslides are dominant. Here, we present results from physical experiments of a single braided channel fed from a fixed inlet and with a falling base-level. The experiments, performed at the St. Anthony Falls Laboratory, provide insights into the formation and distribution of landslides during times of base-level fall. We conducted six experiments, each with a constant water discharge of 0.1 L/s and sediment flux of 0.022 L/s (including pore space) of silica sand with a unimodal grain size distribution (D50 = 0.14mm). We tested base-level fall rates of 0 mm/hr, 25 mm/hr, 50 mm/hr, 200 mm/hr, 300 mm/hr, and 400 mm/hr. Overhead photographs were taken every 20 seconds and digital elevation models (DEMs) were collected every 15 minutes throughout each experiment. We are able to determine the frequency and magnitude of landslides by mapping them throughout each experiment. To obtain the volume of the landslides, we used a combination of the georeferenced images and the DEMs. The magnitude-frequency distribution of landslides in the field has been observed to follow a power-law scaling. We can compare the magnitude-frequency distributions of the experimental landslides and see if they display similar scaling. Knickpoints play an instrumental role in the formation of landslides, but we want to determine, specifically, how the slopes upstream and downstream of the knickzone will influence landslide distribution. Downstream of the knickpoint, the channel is steeper and the valley is deeper. We are investigating how these two factors affect the distribution of landslides, which in turn effects the sedimentation rate in the fluvial system. This additional sediment inhibits river incision and can reshape the river. The analysis and mapping of the landslides will provide insight into: 1) the relationship between base-level fall and landslides, 2) how knickpoint migration alters landslide patterns, and 3) how the lateral movement of the river channel affects landslide distribution.