Rocky Mountain (56th Annual) and Cordilleran (100th Annual) Joint Meeting (May 3–5, 2004)

Paper No. 5
Presentation Time: 9:40 AM

REMOTE SENSING AND FIELD STUDIES OF THE SALMON FALLS LANDSLIDE, SOUTHERN IDAHO


THACKRAY, Glenn D., Department of Geosciences, Idaho State University, Pocatello, ID 83209, DORSCH, Stephen J., Idaho State University, Campus Box 8072, Pocatello, ID 83209-8071, CHADWICK, John, Geosciences, Idaho State Univ, Box 8072, Pocatello, ID 83209 and GLENN, Nancy, Geosciences, Idaho State Univ, 12301 W. Explorer Drive, Suite 102, Boise, ID 83713, thacglen@isu.edu

The Salmon Falls Landslide is a 0.3 km2, canyon-wall landslide in the Salmon Falls Creek canyon near Buhl Idaho. The landslide, which began apparent movement in 1999, has dammed the creek and created a 2-km-long lake, with potential dam-outburst hazards for irrigation equipment and isolated homes downvalley. In order to assess those hazards and determine the motion of the landslide, we have used a variety of remote sensing and ground-based methods.

The landslide is a bearing failure in late Tertiary sediments and basalt. At the site, the ca. 15-m-thick Lucerne School basalt overlies lacustrine and volcaniclastic sediments of the Glenns Ferry Formation. We infer that the failure plane lies in the Glenns Ferry Formation. Extensive headwall fracturing in the Lucerne School basalt has permitted several m to tens of m subsidence of ca. 5 ha of the eastern canyon rim. The descending block has forced several ha of the canyon floor westward. The landslide toe, located adjacent Salmon Falls Creek, has risen several m. The landslide mass exhibits many fractures, most with extensional motion.

We have determined that the landslide motion began prior to 1998 and peaked in 2001. Quantitative comparison of archival aerial photos with 2002 Quickbird imagery yields total lateral movement of 8-16 m between 1990 and 2002. However, only 5-11 m occurred between 1998 and 2002, suggesting that a substantial portion of total movement occurred between 1990 and 1998. Theodolite surveys and autonomous GPS station data indicate that vertical and horizontal motion rates peaked in 2001 at ca. 20 cm/month and have since slowed to < 1 cm/month. GPS data also indicate that movement directions of the landslide toe changed from west to southwest in April, 2003, coincident with the canyon wall block shifting from downward-northwest to downward-only motion. Given current motion rates and pronounced dam armoring, the landslide appears to pose minimal hazard at this time.