ESTIMATING SEDIMENT DELIVERY FROM A LARGE BEDROCK LANDSLIDE

The potential increase in sediment delivery that might result due to a proposed land management practice was estimated from research relating to rainfall, evapotranspiration, and run off; geotechnical parameters, and geologic and geomorphic evidence for a large landslide along the Van Duzen River in Humboldt County, California. The landslide in question, here referred to as the VDR Landslide (VDRLS), is a classic translational-bedrock landslide failing on moderately-steep, north-dipping bedding planes in a fine-grained member, the Carlotta Formation, of the Plio-Pleistocene Wildcat Group. The landslide is about 500 m long by 350 m wide, and estimated to be about 30 to 35 m thick. The VDRLS is nested in a larger landslide that is about twice the size of the VDRLS. The margins of the VDRLS are characterized by well developed system of ground cracks and a scarp/pull away zone that displays roughly parallel bands of exposed bedrock and ages of invading vegetation that record sequential episodes of movement and inactivity. Within the landslide are areas characterized by ground cracks up to several meters wide and deep. Elsewhere within VDRLS horst and graben terrain, 3 to18 m of local relief is common. Segments of the toe of the landslide crop out in the channel of the Van Duzen River and, over time, portions of the channel have been uplifted about 7 m as the river has maintained a low flow channel across the toe of the landslide. Movement may have been triggered by an initial period of intense bank erosion due to a major storm event in 1964. Subsequent movement appears to be the result of down-stream migration of bank erosion.

The potential for increased sediment delivery and the possible volumes over current delivery rates were estimated by evaluating the possible magnitude of increase in the water table within the landslide following the proposed management activities, and then estimating the influence this increase would have on the current factor-of-safety of the landslide (which seasonally exists above or below 1). Then assuming a “reasonable” worse case scenario, the potential sediment delivery was calculated and compared to the annual total sediment load in the local area of the landslide as well as the entire Van Duzen River system.