Rocky Mountain - 54th Annual Meeting (May 7–9, 2002)

Paper No. 0
Presentation Time: 9:40 AM

DEEP-SEATED LANDSLIDE HAZARDS IN THE CENTRAL GRAND CANYON, ARIZONA


SAVAGE, Jill E., Department of Geology and Geological Engineering, Colorado School of Mines, Golden, CO 80401, SAVAGE, William Z., U.S. Geol Survey, Mail Stop 966, Denver Federal Center, Denver, CO 80225 and HUNTOON, Peter W., PO Box 60850, Boulder City, NV 89006, savage@usgs.gov

Rotational landslides have blocked the Colorado River and its tributaries in the central Grand Canyon. These blockages have forced the rivers to reroute around landslide debris and excavate new channels. The landslides are at least 600 m thick, cover surface areas up to 16 km2, and are composed of rocks from the Bright Angel Shale through the Esplanade Sandstone. The slip surface of each landslide consistently soles in the Bright Angel Shale.

The third author has hypothesized that rotational landslides occur shortly after the incision of the Bright Angel Shale. This hypothesis suggests that as the river cuts down through the Bright Angel Shale the canyon profile becomes oversteepened and intensifies the already extreme topographic relief of the canyon walls. As down cutting continues, the structurally weaker Bright Angel Shale can no longer support the load imposed on it by the overlying strata and the canyon walls fail.

We have used a finite-element model to test the hypothesis and to predict the development of deep-seated rotational landslides at two sites in the central Grand Canyon. At the first site, east of Kanab Canyon at Colorado River mile 142, the Bright Angel Shale has not been incised and landslides have not occurred. Here, the finite element modeling predicts observed lateral flow of the Bright Angel Shale towards the river channel, which occurs during incision of the Muav Limestone. Also, the finite element modeling suggests that with further incision into the underlying Bright Angel Shale the northwest canyon wall will become unstable and develop a deep-seated rotational landslide.

The second site is the Deer Creek landslide complex located between Colorado River miles 136 and 138. Here, the Bright Angel Shale is incised and multiple landslides have occurred. Results of finite element modeling show a pattern of failure along the north canyon wall that is in reasonable agreement with the geometry of the Deer Creek landslide complex. The results substantiate the hypothesis of failure occurring after the exhumation of the Bright Angel Shale by the incising river.

We conclude that future landslides and damming of the Colorado River may occur between Colorado River mile 142 and Parashant Canyon, where the Bright Angel Shale is next to be exposed by the incising river.