CHAPTER 4

RETREAT OF THE SLUMGULLION MAIN SCARP

by Robert L. Schuster and William K. Smith

Introduction

The nearly vertical main scarp (fig. 1) of the Slumgullion earth flow is approximately 1,000 m long and 230 m high. This magnificent scarp, which formed in hydrothermally altered Tertiary volcanics, is slowly retreating in response to freeze-thaw activity, sapping, and erosion. The volcanic rock has been weakened by jointing and by numerous faults that add to the instability of the main scarp (Diehl and Schuster, this volume).

Active rock-avalanche chutes are located along main-scarp faults and fault intersections. These chutes funnel rock debris from the main scarp to the talus apron at its base (fig. 2). The talus deposits are made up of angular cobble- to boulder-sized blocks of volcanic rock. The large boulders, some as large as automobiles (fig. 3), are derived from failure of the main scarp along vertical joints in the resistant Hinsdale Formation at the top of the scarp (Diehl and Schuster, this volume). The talus deposits load part of the head of the landslide and potentially serve as a long-term driving force applied to the mass. However, we have little idea of the rate of accumulation of this rock-fall/rock-avalanche material, and it seems to have little effect on the currently active part of the landslide.

Use of Archival Photographs to Estimate Main-Scarp Retreat

Chandler and Cooper (1988) have reviewed methods of using archival aerial or terrestrial photographs to determine morphologic changes of hillside slopes. These methods can be used to compare digital terrain models (DTM's) from stereographic photographs (stereopairs) taken at time intervals in order to quantify a "DTM of difference," i.e., one digital terrain model is "subtracted" from another. The comparison can be used to quantitatively measure landslide movement or slope retreat. Vertical aerial photography, oblique aerial photography, or terrestrial photography can be used, but the photos must give stereo coverage.

High-quality individual terrestrial (i.e., ground-based) photographs of the Slumgullion main scarp taken in 1905 by Whitman Cross of the U.S. Geological Survey are available from the Photo Archives of the U.S. Geological Survey Library, Denver, Colo. Unfortunately, no archival stereopairs of the main scarp are available; thus, quantitative analysis of morphogenetic changes are not possible. Instead, these photographs were replicated by the authors in 1992 and 1993 from the exact sites at which the archival photos were taken. This was done with a modern 35-mm camera and a 35-mm to 105-mm "zoom" lens that could closely approximate the framing of the 1905 photos. By careful comparison of the new photos with the archival photos, it is possible to note local changes in main-scarp geometry, but it is not possible to determine quantitatively the amount or volume of retreat at these points. An example comparing 1905 Cross photography of the main scarp and a photograph taken by the authors in 1992 is presented in figure 4. Comparison of these photos indicates that, since 1905, major changes in surface geometry have occurred in weaker rocks (Diehl and Schuster, this volume) at mid-height on the rock face near the center of the wall (between the arrows in fig. 4A). Little change has taken place in the more-resistant rocks at the top of the wall.

Beginning in 1982, the authors have taken their own terrestrial photographs of the more active parts of the main scarp, avalanche chutes, and talus deposits, especially in the sector of the scarp between the arrows in figure 4A. These photographs are being compared to current photos and will be compared to future photos taken from the same positions to obtain rough estimates of degradation of the most active parts of the scarp. An example of such a comparison over a period of 3 years is presented in figure 5, where differences in distribution of rock-fall boulders from the main scarp can be seen at the toe of the talus apron.

Analytical Use of Stereographic Photographs to Determine Rate of Main-Scarp Retreat

A quantitative approach to determining the rate of retreat of the main scarp can utilize recent developments in analytical photogrammetry that permit monitoring morphologic changes of slopes or vertical faces from stereoscopic ground photos taken with high-quality cameras and calibrated lenses (Chandler and Cooper, 1988; Chandler and Moore, 1989). In 1992, we occupied nine permanent camera stations (fig. 6) about 100 m apart on a line about 1 km in front of the main scarp. From these, a series of stereographic photo pairs (fig. 7) was taken with a calibrated Hasselblad camera. A vertical topographic map of the main scarp as it was in 1992 will be produced using these stereopairs. In future years, the camera stations can be reoccupied, the 1992 stereopairs replicated, and new vertical topographic maps produced. By "subtracting" a new DTM from the 1992 version, the amount of retreat and the volume of failed main scarp can be determined. Because the rate of retreat is very slow, it may be decades before substantial changes will be noted. Thus, our photos will be placed in USGS archives for future use and comparisons.

Conclusions

Rock material that falls, rolls, bounces, or slides from the main scarp has a long-term effect in loading the head of the Slumgullion landslide and, thus, could be a factor in future landslide activity. Probably the most efficient means of determining the amount of material that moves from the main scarp to the head of the landslide is by means of photogrammetric study of the main scarp from terrestrial photographs taken over time spans measured in years or tens of years. Archival terrestrial photos taken by Cross in 1905 can be used to qualitatively determine morphogenic changes in the main scarp, avalanche chutes, and talus deposits. Stereoscopic photography from predetermined ground stations can be used with future replicated stereographic photographs and an analytic plotter or computer analysis to obtain quantitative measures of spatial retreat of the main scarp and deposition of failed rock material onto the head of the landslide.

References Cited

Chandler, Jim, and Cooper, Mike, 1988, Monitoring the development of landslides using archival photography and analytical photogrammetry: Land and Minerals Surveying, Royal Institution of Chartered Surveyors, v. 6, p. 576-584.

Chandler, J.H., and Moore, R., 1989. Analytical photogrammetry: A method for monitoring slope instability: Quarterly Journal of Engineering Geology, v. 22, p. 97-110.